Open-Weight LLMs
Open-source LLMs for private & custom AI
Every model here can be self-hosted and run inside your own environment — the foundation for custom AI applications, operational (ops) automation, and private LLMs where your data never leaves your control.
LLM.co helps enterprise and middle-market teams choose, deploy, and operate these models as part of a private AI Operating System.
597 models
Qwen3-0.6B
A 0.6B reasoning model that runs locally with switchable thinking/non-thinking modes—built for resource-constrained private deployments that need on-demand logic for ops automation and custom workflows.
Qwen3-8B
8B reasoning model with thinking/non-thinking mode toggle—built for private deployment in ops workflows that need cost-efficient reasoning without vendor lock-in.
gpt2
Lightweight, self-hostable text generation base for internal workflows, document automation, and custom fine-tuning—not a chat model, but a foundation for ops-specific AI.
Qwen2.5-7B-Instruct
A 7B instruction-tuned model built for private deployment—strong at coding, math, long-context reasoning, and structured output (JSON)—lets ops teams automate workflows and build custom AI without shipping data to third parties.
Qwen2.5-7B-Instruct
A 7B instruction-tuned model built for private deployment in ops workflows—coding, structured data handling, and multi-language automation at controlled inference cost.
Qwen2.5-1.5B-Instruct
Lightweight instruction-tuned LLM for private ops automation, chatbots, and structured-data workflows on modest hardware.
Qwen3-Embedding-0.6B
Production-grade embedding engine for private document retrieval, search, and RAG—0.6B parameter efficiency meets multilingual scale.
Qwen3-4B
A 4B dense reasoning model that switches between thinking and non-thinking modes—built for ops teams running custom agents, internal automation, and reasoning-heavy workflows entirely on-premise.
DeepSeek-R1
A 671B MoE reasoning model for companies needing complex problem-solving, code generation, and math reasoning in private deployments; trades raw throughput for depth.
Qwen3.6-35B-A3B-NVFP4
Quantized 35B MoE model optimized for private deployment in agent systems, RAG pipelines, and operational automation workflows—3x smaller memory footprint without material accuracy loss.
gpt-oss-20b
A 20B MoE model optimized for low-latency private deployment and custom ops automation with full reasoning chain transparency.
deepseek-v4-gguf
Quantized MoE text generator optimized for private inference on large-RAM systems; DeepSeek V4 Flash compressed into GGUF for operator-controlled deployment.
Qwen3-32B
32B reasoning model with dual thinking/non-thinking modes for private ops automation, agent workflows, and reasoning-intensive custom AI.
Qwen3-1.7B
1.7B reasoning model with thinking/non-thinking mode toggle—sized for private deployment while handling complex ops automation, agent tasks, and custom workflows without external API dependency.
Qwen3-4B-Instruct-2507
A 4B instruction-tuned model for private ops automation and agent deployment—reasoning, tool use, and long-context handling at minimal hardware cost.
Qwen2.5-0.5B-Instruct
Ultra-lightweight instruction-tuned LLM for private deployment in resource-constrained ops environments—chat, structured outputs, and basic automation at minimal compute cost.
dolphin-2.9.1-yi-1.5-34b
Full-parameter fine-tuned 34B conversational & coding model for private deployment—unfiltered, agentic, built for ops automation and custom AI applications where data control matters.
gpt-oss-120b
120B MoE model for private, reasoning-heavy operational AI—automating complex workflows, agent systems, and internal knowledge tasks within a single H100.
Qwen3-14B
A 14B reasoning-capable model with switchable thinking/non-thinking modes, built for private deployment in ops workflows requiring cost-controlled reasoning or fast inference without the overhead.
Qwen2.5-Coder-14B-Instruct
Purpose-built code LLM for private deployment: embed coding agents, code review automation, and developer productivity into your ops stack without vendor lock-in.
Qwen2-1.5B-Instruct
A 1.5B instruction-tuned model sized for edge deployment and private ops automation—fast inference, minimal VRAM, no vendor lock-in.
Qwen2.5-7B-Instruct-AWQ
4-bit quantized 7B instruction-tuned model for cost-efficient private deployment in ops automation, customer support, and knowledge work without data egress.
distilgpt2
Lightweight text-generation backbone for private, on-device operational AI—40% smaller than GPT-2 with Apache 2.0 freedom.
pythia-160m
A 160M research-grade causal LM for building interpretable, fine-tuned custom AI systems that stay on-premises and under your control.
Qwen2.5-32B-Instruct
32B instruction-tuned workhorse for private deployment: strong coding, math, long-context, and structured output generation—built for ops teams automating knowledge work without external API dependency.
Qwen3-30B-A3B
MoE model with dynamic thinking/non-thinking modes for ops teams building private reasoning agents and cost-efficient custom AI without inference overspend.
OTel-LLM-E4B-IT
A 4.5B telecom-specialized LLM fine-tuned for context-grounded RAG pipelines—deploy privately to automate internal telecom ops workflows without exposing domain queries to third parties.
GLM-4.7-Flash
A 30B MoE model purpose-built for efficient private deployment in ops workflows—balances reasoning, coding, and agentic reasoning without the footprint of larger systems.
Qwen3-Embedding-4B
Purpose-built embedding model for building private vector search, retrieval, and ranking systems that keep all text data inside your infrastructure.
Qwen3-Coder-Next-FP8
Sparse-mixture-of-experts coding agent engine for private agentic automation—80B params, 3B active, built for long-horizon ops tasks and local IDE/CLI integration.
SmolLM2-135M-Instruct
Ultra-lightweight instruction-tuned LLM for on-device ops automation, custom workflows, and private deployment on standard edge hardware.
DeepSeek-V4-Flash
A highly efficient 284B-parameter MoE model with only 13B active parameters and 1M-token context—purpose-built for private deployment in ops workflows where long-context understanding and cost-controlled inference matter.
Qwen2.5-14B-Instruct
14B instruction-tuned model for building private, knowledge-rich AI agents and automating complex operational workflows without leaving your infrastructure.
TinyLlama-1.1B-Chat-v1.0
A 1.1B parameter chat model built for private deployment in resource-constrained environments—designed to run customer operational workflows on-premise without external API dependency.
Qwen3-Reranker-0.6B
Purpose-built reranker for retrieval pipelines: rank and filter candidate documents to surface the most relevant results—critical for RAG, search, and agent knowledge systems running privately.
Qwen2.5-0.5B
Ultra-lightweight base LLM (0.5B params) for ops teams building private, embedded AI agents and custom workflows on edge hardware or isolated infrastructure.
Gemma-4-26B-A4B-NVFP4
Quantized 26B MoE model for private, GPU-efficient reasoning and multimodal ops automation—reasoning, coding, function-calling, and agent workflows on customer hardware.
Qwen2.5-Coder-7B-Instruct
Code-specialized 7B instruction-tuned model for private code automation, agent development, and internal coding workflows — sized for on-premise deployment without vendor lock-in.
Qwen2.5-Coder-7B-Instruct
A 7B instruction-tuned code LLM built for private deployment in ops workflows: code generation, agent automation, and internal knowledge tasks without vendor lock-in.
GLM-5-FP8
744B sparse MoE model built for agentic reasoning and long-horizon ops automation—code, systems engineering, and multi-turn workflow tasks at scale.
Qwen3-Reranker-4B
A 4B text-reranking model for private RAG, search ranking, and document relevance automation in enterprise ops workflows.
GLM-5.2-FP8
A 753B open-weight MoE model with 1M context, strong coding/agentic performance, and MIT licensing—built for companies that need long-horizon reasoning, code automation, and private deployment control.
DeepSeek-V3.2
685B parameter open-weight reasoning model for private deployment—built for ops teams automating complex workflows, agentic tasks, and internal knowledge work without cloud dependency.
DeepSeek-R1-0528
Reasoning-intensive LLM for operational automation of analytical, code, and mathematical tasks in private environments
Qwen3-14B-AWQ
14B thinking + non-thinking LLM for private ops automation, agent workflows, and custom reasoning applications.
Qwen2.5-Coder-32B-Instruct-AWQ
A 32B code-specialized LLM optimized for private deployment—generate, reason about, and fix code while keeping training data and models entirely within your infrastructure.
Qwen3-0.6B-FP8
Lightweight reasoning model (0.6B) designed for on-premises deployment where companies need cost-effective, controllable AI for operational automation without cloud dependency.
PowerMoE-3b
Sparse mixture-of-experts model for cost-efficient private deployment—activates only 800M of 3.3B params per token, cutting inference cost and latency for internal ops automation and custom AI without sacrificing quality.
moondream2
A lightweight vision-language model for automating image understanding tasks in private, self-hosted ops environments—captioning, visual Q&A, object detection, and UI element localization without external API calls.
Qwen3-32B-AWQ
A 32B reasoning-capable model with dual thinking/non-thinking modes, quantized to 4-bit AWQ for cost-efficient private deployment in custom AI and operational automation workflows.
Qwen3-Coder-30B-A3B-Instruct
A 30B MoE coding LLM for ops teams building agent-driven automation and custom development tools that stay entirely in their own environment.
diffusiongemma-26B-A4B-it-NVFP4
Multimodal ops engine for document processing, video analysis, and structured reasoning—small active parameter footprint (3.8B) with enterprise quantization for private data handling.
DeepSeek-R1-0528-Qwen3-8B
8B reasoning model distilled from DeepSeek-R1-0528: reasoning-grade math/code performance at small-model cost, deployable privately on modest hardware.
SmolLM2-135M
A 135M parameter instruction-tuned model for on-device private automation, custom task adaptation, and cost-controlled operational AI without cloud dependencies.
Qwen2.5-14B-Instruct-AWQ
A 14B instruction-tuned, 4-bit quantized model optimized for private deployment—strong at coding, math, long-context workflows, and structured JSON output in cost-controlled ops environments.
Qwen3-Coder-30B-A3B-Instruct-FP8
A 30B MoE coding model for private deployment—designed for agentic code generation, repository-scale understanding, and tool-calling automation within your own infrastructure.
Qwen3-VL-30B-A3B-Instruct-AWQ
Vision-language model for private deployment: automate document/image workflows, GUI automation, and multimodal reasoning entirely within your infrastructure.
Qwen2.5-Coder-32B-Instruct
Code-generation and code-reasoning backbone for building private AI agents, internal developer tools, and ops automation that runs entirely in your infrastructure.
Phi-3.5-vision-instruct
Lightweight multimodal model (4.1B params) for private, self-hosted vision+text automation in resource-constrained ops environments.
Qwen2-0.5B
Lightweight base LLM for private deployment in ops automation, custom fine-tuning, and edge inference where data residency and cost control matter more than frontier capability.
Qwen3-Coder-Next
A 3B-activated, 80B-parameter mixture-of-experts coding LLM optimized for agentic automation, tool use, and private IDE integration—built to reduce inference cost while maintaining capability for enterprise coding workflows.
Qwen3-30B-A3B-Instruct-2507
Efficient 30B MoE model for private-hosted ops AI: long-context reasoning, tool calling, and custom workflow automation with controlled data.
gpt2-large
Lightweight, proven text-generation foundation for building private AI workflows and custom language tasks without cloud dependency.
DeepSeek-V4-Pro
A 1.6T-parameter MoE model with 49B activated params and 1M token context, optimized for private deployment in ops workflows requiring long-context reasoning, code generation, and knowledge tasks without external API dependency.
h2ovl-mississippi-800m
Compact vision-language model for OCR, document extraction, and multi-modal ops automation in resource-constrained environments.
TinyLlama-1.1B-Chat-v0.3-GPTQ
A 1.1B chat model in GPTQ quantization—purpose-built for resource-constrained private deployment where ops teams need sub-2GB inference footprint and full data control.
h2ovl-mississippi-2b
Compact vision-language model (2B parameters) for private document AI, OCR, and visual reasoning—runs on modest hardware while staying in your infrastructure.
Qwen2.5-1.5B
Lightweight base LLM (1.5B params, 32K context) for building private, custom AI agents and operational automation without heavy infrastructure.
Mistral-7B-Instruct-v0.2
Compact instruction-tuned model for private deployment in operational AI workflows—support automation, document processing, and internal agent tasks without external API dependency.
Mistral-7B-Instruct-v0.2
Lightweight instruction-tuned workhorse for private, self-hosted deployment in ops automation and custom AI workflows where data residency matters.
SmolLM-1.7B-Instruct-quantized.w4a16
A 1.7B lightweight instruction-tuned model optimized for private deployment in ops workflows—quantized to 75% smaller footprint for edge/on-prem inference without sacrificing reasoning.
Qwen2.5-Coder-14B-Instruct-AWQ
Code-generation and reasoning engine for private ops automation—handling code-heavy workflows, agent scaffolding, and custom development tasks without external API dependencies.
Qwen3-0.6B-Base
Lightweight dense base model (0.6B) for private deployment in ops automation, internal knowledge systems, and custom AI agents where data residency and cost-per-inference matter more than frontier capabilities.
Qwen3-Reranker-8B
Specialized reranking engine for private document retrieval systems—rank search results, agent knowledge bases, and retrieval-augmented pipelines without vendor lock-in.
Phi-3.5-mini-instruct
Compact, reasoning-focused model (3.8B params) for self-hosted ops automation, private document processing, and custom AI agents in memory/latency-constrained environments.
Phi-3.5-mini-instruct
Lightweight reasoning engine (3.8B params) for private ops automation, long-context document workflows, and custom AI in memory/latency-constrained environments.
gte-Qwen2-1.5B-instruct
A 1.5B multilingual text-embedding model for semantic search, retrieval, and similarity tasks—purpose-built for private deployment in ops workflows requiring compact, controllable inference.
OTel-LLM-8B-A1B-IT
Telecom-domain LLM for private RAG pipelines: ground operational Q&A on your own 3GPP/O-RAN/GSMA knowledge without sending telecom data to third-party APIs.
Qwen2.5-1.5B-quantized.w8a8
Lightweight 1.5B quantized assistant for private, resource-constrained ops automation and embedded custom AI without cloud dependency.
DeepSeek-V3-0324
A 685B parameter open-weight reasoning model for private, domain-specific AI applications—reasoning, code generation, and multi-turn workflows that stay within your infrastructure.
Qwen3-4B-Instruct-2507-FP8
A 4B parameter instruction-tuned model optimized for private deployment in resource-constrained ops environments—reasoning, tool use, and long-context tasks without leaving your infrastructure.
Qwen3-235B-A22B
235B MoE model with switchable reasoning modes for private deployment—designed for ops teams building custom reasoning agents and automating complex workflows while keeping data in-house.
Phi-tiny-MoE-instruct
A 3.8B-parameter MoE model (1.1B active) designed for memory-constrained, latency-sensitive operational AI and private deployments where inference cost and response speed matter.
phi-4
14B dense transformer for reasoning, math, and code—sized for latency-bound ops automation and private deployment without sacrificing capability.
Mistral-7B-v0.1
A lean, inference-optimized 7B base model for private-deployment ops AI and custom applications where data residency and control matter more than frontier capability.
gemma-4-31B-it-assistant
A 31B multimodal dense model designed for private deployment across text, vision, and reasoning tasks—with speculative decoding to cut inference latency while keeping data in your environment.
pythia-70m-deduped
Research-grade 70M base model for building lightweight, interpretable custom AI applications and automating document/text workflows in a fully private environment.
DeepSeek-R1-Distill-Qwen-32B
Dense reasoning distill for private ops: chain-of-thought problem-solving at 32B scale, built to run on enterprise hardware without vendor dependency.
phi-2
A lightweight 2.7B base model for building private QA, chat, and code-assist agents without external API dependency.
GLM-5.1-FP8
A 754B agentic/code-focused LLM optimized for long-horizon reasoning and tool use—built for ops teams to run as a private, controllable backbone for automation agents and custom applications.
DeepSeek-R1-0528-NVFP4-v2
FP4-quantized 393B reasoning model for private deployment—cut VRAM needs ~40% vs FP8, maintain high reasoning accuracy, run on your own GPUs.
Qwen2.5-32B-Instruct-AWQ
A 32B instruction-tuned LLM optimized for private deployment with strong coding, math, and structured output capabilities—built for ops teams automating internal workflows while maintaining data control.
Qwen2.5-1.5B-Instruct-AWQ
Lightweight instruction-tuned LLM (1.5B, 4-bit AWQ) for private deployment in ops workflows—chat, structured output, coding tasks—without GPU bloat.
Qwen2.5-Coder-1.5B-Instruct
A 1.5B code-generation specialist for embedding in private ops workflows—code agents, internal tooling automation, and custom AI layers that stay in your environment.
indic-parler-tts
Private text-to-speech for Indic languages: run multilingual voice synthesis in-house, keeping audio generation and user voice data fully under your control.
SmolLM3-3B
3B reasoning model designed for private deployment in ops workflows—reasoning, tool-calling, and multilingual automation without cloud dependencies.
Qwen2-7B-Instruct
A 7B instruction-tuned model built for private deployment as the reasoning backbone in operational AI workflows—coding, document processing, and internal agent logic.
gemma-4-12B-coder-fable5-composer2.5-v1-GGUF
A 12B Python coding model distilled on verified chain-of-thought reasoning, sized to run entirely on-premises with minimal hardware footprint—purpose-built for teams automating code review, documentation, and structured problem-solving in isolated environments.
Qwen2.5-72B-Instruct
72B instruction-tuned model for private deployment—coding, math, long-context reasoning, and structured output generation in controlled environments.
granite-4.1-8b
8B instruct model built for private deployment, tool-calling agents, and ops automation—control your own inference without vendor lock-in.
Bielik-11B-v3.0-Instruct-awq
Multilingual European language model optimized for Polish—AWQ-quantized for efficient private deployment in ops workflows across 32 languages.
Mixtral-8x7B-Instruct-v0.1
Sparse Mixture-of-Experts model for building private, instruction-following AI agents and custom ops automation without vendor lock-in.
DeepSeek-R1-Distill-Qwen-1.5B
Distilled reasoning model (1.5B) for private ops automation: math, logic, and structured problem-solving tasks that fit on a single machine.
DeepSeek-R1-Distill-Llama-70B
A 70B dense reasoning model distilled from DeepSeek-R1, optimized for math, code, and complex logic tasks—deployable entirely on-premise to keep proprietary reasoning workflows private.
Qwen3-1.7B-Base
Lightweight dense LLM (1.7B) for private deployment in ops workflows: multilingual reasoning, code, and STEM tasks without heavy infrastructure or vendor lock-in.
Qwen3-30B-A3B-Instruct-2507-AWQ-4bit
Production-grade 30B sparse MoE model for private, long-context AI workloads: instruction-following, reasoning, coding, and agentic automation in controlled environments.
Qwen3-8B-FP8
8B dense reasoning model with thinking/non-thinking modes for private deployment—custom ops AI, agent workflows, and internal knowledge automation without leaving your infrastructure.
Phi-3-mini-4k-instruct
Lightweight 3.8B model for building reasoning-heavy custom AI and automating ops workflows in resource-constrained or latency-critical environments, fully self-hosted.
Qwen2.5-7B-Instruct-bnb-4bit
A quantized 7B instruction-tuned LLM optimized for cost-efficient private deployment and fine-tuning in ops workflows requiring multilingual support and structured output generation.
tiny-random-Llama-3
Tiny Llama 3 derivative for rapid prototyping of private, lightweight text-generation workflows without enterprise compute overhead.
bge-reranker-v2-gemma
A multilingual reranker for RAG systems and retrieval pipelines—rank passages by relevance to queries, run it privately.
Qwen3-4B-Base
Lightweight multilingual base model for private custom AI and operational automation in resource-constrained enterprise environments.
Rio-3.0-Open-Mini
A 4B distilled reasoning model optimized for private deployment in ops workflows requiring math, code, and structured problem-solving without external API dependency.
DeepSeek-V4-Flash-NVFP4
High-performance quantized MoE reasoning model for enterprises running private AI agents, agentic workflows, and custom applications on NVIDIA infrastructure.
Qwen3-4B-AWQ
A 4B quantized reasoning model that runs on modest hardware while supporting both thinking (chain-of-thought) and fast inference modes—purpose-built for private ops automation and custom AI applications.
Hermes-4-14B-AWQ-4bit
14B reasoning model for private ops automation: structured outputs, tool calling, math/code tasks, and steerable alignment—built to stay in your environment.
VLM2Vec-Full
Multimodal embedding model for private, in-house search, retrieval, and similarity tasks across images and text—built on Phi-3.5-V, deployable entirely on your infrastructure.
Qwen3-Embedding-4B-W4A16-G128
Lightweight embedding model for private semantic search, RAG, and vector operations—quantized to fit on modest ops infrastructure.
Qwen3-Reranker-4B-W4A16-G128
A quantized 4B reranker for private ranking/relevance pipelines—compress your search and retrieval ops without the memory footprint.
Phi-4-mini-instruct
3.8B instruction-tuned model for memory-constrained private deployments requiring reasoning and multilingual support without external API dependencies.
Qwen3.6-27B-NVFP4
A quantized 27B reasoning model optimized for private deployment in agent systems, RAG, and operational automation—4-bit precision cuts GPU memory ~2.5x without sacrificing accuracy.
Qwen2.5-Coder-7B-Instruct-GPTQ-Int4
Code-generation engine for private, custom AI agents and automation workflows—built small enough to self-host, sharp enough to handle real development tasks.
Phi-4-multimodal-instruct
A lightweight multimodal foundation model (5.6B params) designed for memory-constrained ops environments that need unified text, vision, and audio processing in a single forward pass—enabling private deployment of complex AI workflows without model chaining.
Qwen3-Coder-480B-A35B-Instruct-FP8
A 480B mixture-of-experts code model for enterprises building agentic coding systems, repository-scale automation, and tool-integrated workflows—deployable entirely on-premise.
Qwen3-4B-Thinking-2507
A 4B reasoning-focused LLM for private-hosted ops automation and agentic workflows where internal data stays in-house and cost-per-inference matters.
Qwen2-0.5B-Instruct
Ultra-lightweight instruction-tuned LLM for private deployment in resource-constrained ops environments—chat, document automation, and agent backbone without cloud dependency.
Qwen2.5-Coder-7B
Purpose-built code generation and reasoning engine for private, self-hosted AI systems that need to automate code-centric operational workflows and integrate coding intelligence into internal tools.
Ornith-1.0-35B-GGUF
Purpose-built coding agent for private deployment—automate development ops, code review, and software engineering workflows without external API dependencies.
Qwen2.5-7B
A 7B base model purpose-built for private deployment and fine-tuning in operational AI workflows—coding, math, structured data handling, and long-context tasks.
xlnet-base-cased
Encoder-only fine-tuning backbone for internal NLP classification, ranking, and Q&A—not a generative model; purpose-built for ops workflows that extract meaning from documents, support tickets, and knowledge bases.
MobileLLaMA-1.4B-Chat
Lightweight instruction-tuned LLM for private deployment on resource-constrained infrastructure; operationally suitable for on-device or edge automation without cloud dependency.
Qwen3.5-397B-A17B-NVFP4
A 397B MoE model quantized to FP4 for cost-efficient private deployment in ops automation, RAG, and agentic workflows on NVIDIA infrastructure.
gpt-neox-20b
Base model for fine-tuning custom ops workflows and internal knowledge agents; viable for private deployment where data residency is non-negotiable.
Qwen2.5-32B-Instruct-GPTQ-Int4
A 32B instruction-tuned model optimized for private deployment, structured outputs, and long-context ops workflows—quantized to run on modest VRAM.
MiMo-7B-RL
7B reasoning-optimized model for private deployment in ops automation, math/code workflows, and custom AI agents—built from scratch for inference efficiency without sacrificing problem-solving depth.
Qwen2.5-Math-1.5B-Instruct
Specialized mathematical reasoning engine for private deployment—solves algebra, calculus, and symbolic problems via chain-of-thought or tool-integrated reasoning without touching external APIs.
Qwen3-30B-A3B-Thinking-2507-AWQ-4bit
A 30B MoE reasoning model (8-of-128 experts active) optimized for complex operational workflows, agentic automation, and self-hosted private AI—trading latency for depth-of-thought on reasoning-heavy tasks.
Ornith-1.0-9B-GGUF
Agentic coding model optimized for private deployment—use it to automate code generation, repository navigation, and software engineering workflows within your infrastructure.
Qwen3-30B-A3B-FP8
MoE reasoning engine for private ops automation—thinking and non-thinking modes in one model, deployable entirely on-premise with native 32K context and YaRN extension to 131K.
GLM-5.2-NVFP4
NVIDIA's quantized GLM-5.2 (NVFP4): a 753B MoE model compressed to 4-bit for on-premises reasoning agents, long-context ops workflows, and tool-use automation without data leaving your infrastructure.
gpt-neo-125m
Lightweight private-deployment text generator for operational automation and custom AI prototyping where data residency and control matter more than frontier capability.
macbert4csc-base-chinese
A specialized Chinese spelling-correction model (BERT-based) for automating text-quality tasks in ops workflows—designed to catch character-level typos and semantic errors at scale.
Qwen3-8B-Base
A 8.2B dense base model for private, multilingual ops automation and custom AI applications requiring 32K context and broad language reasoning.
Bielik-11B-v3.0-Instruct
Multilingual 11B instruction-tuned model for private-deployment ops automation and custom AI in non-English enterprises.
DeepSeek-V3.2-AWQ
685B parameter reasoning model optimized for private deployment via quantization (AWQ 4-bit), enabling companies to run sophisticated agentic AI and complex problem-solving workflows entirely within their own infrastructure.
gpt-oss-20b-GGUF
A 20B MoE model for private deployment, fine-tuning, and agentic automation within resource-constrained ops environments.
Qwen3-30B-A3B-abliterated
Uncensored 30B MoE model for private deployment in ops workflows where instruction-following and unrestricted generation matter.
GLM-5.2-GGUF
Long-context reasoning and coding model optimized for private deployment via GGUF quantization, with 1M-token window and flexible inference cost/quality tradeoffs.
GLM-4.5-Air
Compact hybrid-reasoning MoE for private agent automation and operational workflows—12B active parameters, MIT-licensed, runs cost-effectively on standard enterprise GPU infrastructure.
GLM-4.5-Air-AWQ-4bit
Compact hybrid-reasoning MoE model for private agent automation and operational AI—12B active parameters, MIT-licensed, production-ready for self-hosted deployments.
Zamba2-1.2B-instruct
Lightweight hybrid SSM-transformer for private deployment in ops workflows—instruction-tuned, low-latency inference, minimal memory footprint.
Qwen3-Next-80B-A3B-Instruct-FP8
80B sparse MoE model built for private, long-context ops automation—reasoning, coding, and agentic workflows at 3B active parameters per token.
Qwen3-8B-AWQ
8B dense model with switchable thinking mode—built for ops teams to run private reasoning workflows and multi-turn automation without external API dependency.
dots.mocr
A 3B-parameter multimodal OCR engine that converts documents, charts, tables, and structured graphics into machine-readable formats (text, markdown, SVG)—designed for ops teams automating document processing pipelines in private environments.
mistral-7b-v0.3-bnb-4bit
A quantized 7B instruction-tuned base for rapid private fine-tuning and operational automation without GPU infrastructure bloat.
gemma-4-12B-agentic-fable5-composer2.5-v2-3.5x-tau2-GGUF
A 12B coding + agentic model optimized for private tool-use, terminal automation, and multi-step technical workflows—purpose-built to run locally with minimal VRAM and no cloud dependency.
gpt2-medium
Lightweight base model for building private text-generation agents and operational automation without cloud dependencies.
Qwen3-4B-GGUF
A 4B dense model with native thinking/non-thinking modes, built for companies running private reasoning workflows and operational automation without external API dependency.
Qwen-AgentWorld-35B-A3B-GGUF
Purpose-built language world model for simulating agent interactions across 7 operational domains (tools, search, terminal, software engineering, mobile, web, OS)—deploy privately to automate complex multi-step workflows without external API dependency.
gpt-oss-20b-MXFP4-Q8
A 20B quantized base model for private deployment and operational automation in resource-constrained environments—optimized for companies needing on-device inference without cloud dependencies.
dots.ocr
Unified document parser that replaces multi-model pipelines with a single VLM for layout detection, text extraction, table parsing, and formula recognition—enabling ops teams to automate document workflows and reduce infrastructure complexity in self-hosted environments.
MiniCPM5-1B
A 1B dense transformer for on-device reasoning, tool-use agents, and code generation—built to run in constrained environments while staying under your control.
gemma-4-26B-A4B-it-assistant
A 26B MoE model optimized for fast private inference on ops tasks—reasoning, document processing, and agentic workflows—with 256K context and multimodal I/O.
Qwen3.6-27B-Text-NVFP4-MTP
Text-only 27B quantized LLM optimized for private, high-throughput deployment on Blackwell GPUs—speculative decoding built-in, data stays in your environment.
LLaDA-8B-Instruct
8B diffusion-based instruction-tuned model for private deployment in ops automation and custom AI applications without third-party inference costs.
DeepSeek-R1-Distill-Llama-8B
Distilled reasoning model for ops teams building private, cost-efficient AI agents that need strong math/code/logic without 671B infrastructure.
gemma-4-E4B-it-assistant
Efficient multimodal instruction-tuned model for on-device and private operational AI—text, image, and audio reasoning in 4.5B effective parameters with speculative decoding for low-latency workflows.
Qwen3-Embedding-4B
Purpose-built embedding model for enterprises to embed company documents, queries, and code into searchable vectors—deployable fully on-premises for private RAG, search, and retrieval automation.
granite-4.0-h-small
32B instruction-tuned model designed for enterprise automation, function-calling, and RAG workflows — built for self-hosted deployment in ops-heavy organizations.
GLM-4.7-Flash
30B MoE model for ops teams that need fast, accurate inference on private infrastructure without cloud dependency.
Mistral-7B-Instruct-v0.1
7B instruction-tuned model for private, self-hosted conversational AI and operational task automation in mid-market environments.
granite-4.1-3b
3B instruction-tuned model for private ops automation, tool-calling workflows, and embedded AI agents in data-sensitive environments.
DeepSeek-R1-0528-Qwen3-8B-MLX-4bit
4-bit quantized 8B reasoning model optimized for Apple Silicon—private, on-device inference for operational AI without cloud dependencies.
Jan-v3.5-4B-gguf
A 4B parameter conversational model with built-in personality and math reasoning, optimized for private deployment and fine-tuning into custom operational workflows.
DeepSeek-R1-Distill-Qwen-7B
A distilled, reasoning-capable 7B model for private reasoning workflows—math, code, support automation—without API dependencies or data leaving your infrastructure.
Qwopus3.6-27B-Coder-MTP-GGUF
27B agentic coder for private deployment: reasoning + tool-use + repo-level tasks, built for ops teams automating code workflows without external API dependency.
pythia-410m
A lightweight, research-grade causal language model for building interpretable custom AI agents and automating text-generation workflows on modest hardware—designed for ops teams who need to own their model weights and data.
Intern-S1-Pro
Trillion-parameter MoE multimodal model for private scientific reasoning, custom AI agents, and enterprise ops automation with full data control.
Qwen3-30B-A3B-Instruct-2507-FP8
A 30B MoE instruct model (3.3B active params) with 262K context and FP8 quantization, built for private deployment in ops workflows—reasoning, tool-calling, and long-document automation.
Qwen2.5-14B-Instruct
14B instruction-tuned LLM for private deployment into ops workflows—coding, structured data, long-context document processing, and multilingual customer/internal automation.
rank1-7b
A reasoning-based reranker for private retrieval systems: make relevance judgments transparent and defensible by generating explicit reasoning chains before scoring documents.
DeepSeek-R1-0528-Qwen3-8B-MLX-8bit
8-bit MLX-quantized reasoning model for Apple Silicon—run DeepSeek-R1 locally without cloud dependency, ideal for private ops automation and custom AI on resource-constrained infrastructure.
Qwen3-14B-FP8
A 14B dense reasoning model with thinking/non-thinking mode toggle, built for private deployment where ops teams need controllable inference cost and transparent chain-of-thought for complex workflows.
japanese-gpt-neox-small
Lightweight Japanese text generation engine for private, custom workflows—customer-controlled LLM for enterprise Japanese language automation.
t5-3b
Text-to-text encoder-decoder for operational NLP tasks: summarization, translation, Q&A, classification — run privately, fine-tune on proprietary data, control outputs entirely.
GLM-5.2
753B sparse MoE model built for long-context reasoning, coding, and agentic workflows—deployable privately with 1M token context and open MIT licensing.
Qwen2.5-Coder-7B-Instruct-AWQ
A 7B code-focused LLM optimized for private deployment, agent automation, and custom coding workflows in mid-market ops environments.
GLM-4.7-Flash-MLX-8bit
A lightweight 30B MoE model optimized for private Apple Silicon deployment, targeting operational AI and custom automations where data residency and inference latency matter.
Ornith-1.0-35B
Agentic coding model optimized for self-hosted deployment; built to automate software engineering tasks—code generation, debugging, repository navigation—within your own infrastructure.
Mistral-7B-Instruct-v0.2-AWQ
Compact 7B instruction-tuned model optimized for low-latency inference on consumer/edge GPU hardware—ideal for ops teams automating internal workflows without external API dependency.
DeepSeek-V3.1
671B MoE model for private deployment: thinking + non-thinking modes, tool-calling, and code agents—built for ops teams running inference in-house.
SmolLM2-360M-Instruct
Lightweight instruction-tuned model (360M params) for on-device private AI and ops automation where data residency and resource constraints matter.
gpt-j-6b
A 6B-parameter causal language model for private text generation and custom AI fine-tuning—runs fully self-hosted, no API dependency.
Qwen3-Coder-30B-A3B-Instruct-AWQ
A 30B Mixture-of-Experts coding LLM optimized for agentic automation, long-context code repository understanding, and private deployment with explicit 4-bit quantization tradeoffs.
granite-4.0-h-tiny
A 7B instruction-tuned model built for enterprise ops automation, tool-calling, and private deployment in mid-market workflows.
Phi-3-vision-128k-instruct
Lightweight multimodal model (4.1B params, 128K context) for private document analysis, process automation, and vision-driven operational workflows in resource-constrained environments.
wildguard
Safety classifier for private AI systems: catch unsafe outputs before they reach users, built into your ops stack.
Qwen3-14B-unsloth-bnb-4bit
A 14B reasoning-capable dense model with thinking/non-thinking mode toggle—purpose-built for private ops AI that needs verifiable logic chains without the overhead of larger MoE systems.
llama-160m
Lightweight speculative inference base model for companies building cost-efficient private LLM systems and internal automation workflows.
Phi-3-mini-128k-instruct
A 3.8B parameter instruction-tuned model built for memory-constrained, latency-sensitive ops environments where reasoning (code, math, logic) and 128K context matter.
DeepSeek-V3.2-Exp
685B sparse-attention model for long-context private inference—built for ops teams running cost-conscious reasoning workloads and custom AI without cloud dependencies.
Qwen3-32B-NVFP4
FP4-quantized 32B model optimized for private deployment on NVIDIA GPUs—ready to power internal agents, RAG, and workflow automation without external API calls.
Qwen3-Coder-30B-A3B-Instruct-GGUF
A 30B sparse-MoE coding model designed for agentic automation and long-context repository understanding in self-hosted ops-AI systems.
Qwen3-Next-80B-A3B-Instruct
Sparse 80B MoE instruct model for private, cost-efficient ops automation and long-context agentic workflows—3B active params, 256K native context, designed for self-hosted deployment.
Qwen2.5-Coder-1.5B
Lightweight code-specialized LLM (1.5B) for private, on-device code generation, reasoning, and fixing in ops workflows—fast enough to embed in internal tools without GPU clusters.
GLM-4.7-Flash-MLX-6bit
A 30B-parameter conversational model quantized to 6-bit for Apple Silicon, designed to run locally in private ops stacks without external API dependency.
MiMo-7B-Base
A 7B reasoning model optimized for math, code, and complex problem-solving tasks in private deployments where companies need lightweight, controllable inference with strong reasoning performance.
Qwen3-Coder-30B-A3B-Instruct-AWQ-4bit
MoE coding model (30B params, 3.3B active) optimized for agentic code generation, tool calling, and repository-scale understanding in self-hosted environments.
DeepSeek-V4-Flash-GGUF
GGUF-quantized inference engine for running DeepSeek's fastest model in-house with minimal VRAM overhead.
Olmo-3-7B-Instruct-SFT
7B instruction-tuned model for companies building private, operator-controlled AI agents and custom workflows that demand reasoning over math, code, and multi-step tasks.
DialoGPT-medium
A lightweight, conversational response model for building private chatbots and dialogue agents that run entirely on your infrastructure.
Qwen2.5-1.5B-Instruct-GGUF
Lightweight instruction-tuned model (1.5B) for private deployment in resource-constrained ops environments—coding, JSON generation, and multi-step workflows without external API dependency.
Qwen3-Coder-Next-GGUF
Mixture-of-Experts coding model optimized for agentic automation, tool-use, and private deployment in ops workflows.
Qwen3-30B-A3B-Thinking-2507
A 30B MoE reasoning model for private deployment—excels at complex ops automation (logic, coding, agent workflows) where thinking-mode inference stays within your firewall.
Mistral-Small-24B-Instruct-2501-AWQ
Quantized 24B instruction-tuned model for private ops workflows, agentic automation, and cost-effective custom AI—deployed entirely in your environment.
Qwen3-235B-A22B-Instruct-2507-FP8
A 235B mixture-of-experts model (22B active) for ops teams building private AI agents, document automation, and reasoning-heavy internal workflows at scale.
openai-gpt
GPT-1: a lightweight, MIT-licensed baseline transformer for private text generation and fine-tuning on operational language tasks.
pythia-6.9b
Research-grade 6.9B causal LM for building interpretable, self-hosted language agents and operational automation without downstream fine-tuning overhead.
PowerLM-3b
A 3B parameter model optimized for small-footprint private deployment across support automation, document processing, and operational reasoning tasks.
Qwen3-4B-Thinking-2507-FP8
A 4B reasoning-focused model designed for private deployment in ops workflows—extended thinking + 262K context for complex automation, compliance analysis, and decision support without data leaving your infrastructure.
Qwen2.5-Math-1.5B
Specialized math reasoning engine for private deployment: solves English/Chinese math problems via chain-of-thought and tool-integrated reasoning, sized for edge/on-prem ops.
Qwen2-1.5B
A compact 1.5B base model for private deployment in ops workflows—coding, math reasoning, multilingual tasks—where data stays in-house and model overhead is minimal.
Qwen3-14B-GPTQ-Int4
A 14B quantized reasoning model for private ops automation—thinking and non-thinking modes in one, built to run on modest hardware while handling complex logic.
Phi-mini-MoE-instruct
Lightweight MoE model (2.4B active / 7.6B total params) for cost-efficient private deployment in ops automation, internal agents, and latency-sensitive custom AI without sacrificing reasoning quality.
Hermes-4-14B
14B frontier reasoning model for private deployment—math, code, logic, tool use, and structured output automation without external API dependency.
zephyr-7b-beta
A 7B chat model fine-tuned for instruction-following and conversational tasks—lightweight enough to run private, capable enough to power custom support agents, internal assistants, and operational automation without external API calls.
zephyr-7b-beta
A 7B chat model fine-tuned for conversational tasks—deployable in private infrastructure to automate support, document triage, and operational Q&A without data leaving your environment.
LightOnOCR-1B-1025
End-to-end document OCR & parsing for ops teams: extract structured text from PDFs, receipts, forms, and tables at scale—self-hosted, no external pipeline needed.
pythia-1b
A lightweight, interpretability-focused base model for building custom operational AI systems and private language applications without heavy infrastructure.
Qwen2.5-3B-Instruct-unsloth-bnb-4bit
Lightweight 3B instruction-tuned model optimized for private deployment and fine-tuning on constrained hardware; 4-bit quantized for on-premises operational AI without external API calls.
VibeVoice-1.5B
Text-to-speech engine for generating long-form, multi-speaker conversational audio (podcasts, dialogue) as a private, controllable foundation for custom audio-generation workflows.
gemma-4-31B-it-NVFP4-turbo
A quantized 31B instruction-tuned LLM optimized for high-throughput, low-latency private deployment on Blackwell GPUs—built for ops teams automating workflows at scale without leaving your infrastructure.
Karnak-40B-v1.0
Bilingual (Arabic–English) 40B depth-extended model for private deployment and custom fine-tuning in Arabic-heavy operational workflows.
Qwen3Guard-Gen-0.6B
A 0.6B safety classification model purpose-built for real-time content moderation in private, self-hosted LLM pipelines—classify prompts and responses as Safe/Unsafe/Controversial across 119 languages.
Qwen3.6-35B-A3B-DFlash
A lightweight draft model for speculative decoding that accelerates Qwen 3.6-35B inference 2–3.6x by proposing tokens in parallel, designed for high-throughput private deployments where latency and cost per inference matter.
Qwen3-Coder-30B-A3B-Instruct-MLX-4bit
Apple Silicon–optimized coding assistant for private deployment in ops workflows—automate code review, doc generation, and internal tooling without external API calls.
SmolLM-135M
Lightweight, self-hostable text-generation engine for lean ops automation and embedded custom AI within resource-constrained enterprise environments.
OLMo-2-0425-1B
A compact 1B base model for private-deployment ops AI and custom fine-tuning in resource-constrained environments.
Qwen3-Embedding-8B-AWQ-INT4
Quantized embedding model for private semantic search and RAG—run locally on modest hardware without cloud dependency.
mzansilm-125m
A lightweight, multilingual decoder-only LM trained on South African languages—designed for private deployment and fine-tuning in low-resource, localized NLP workflows.
step3
A 321B MoE vision-language model for private, cost-efficient multimodal reasoning in custom AI workflows—38B active parameters per token minimize compute overhead in self-hosted deployments.
Step-3.5-Flash
Sparse MoE reasoning engine for private deployment—11B active parameters, 256K context, purpose-built for agentic workflows and long-horizon operational tasks.
voyage-4-nano
A 346M-parameter multilingual embedding model for private semantic search, RAG, and retrieval automation—engineered for on-premise deployment without indexing friction.
Qwen2.5-Math-7B-Instruct
Specialized math-reasoning LLM for private deployment in ops workflows requiring symbolic computation, equation solving, and chain-of-thought verification.
Phi-3.5-MoE-instruct
A 41B-parameter mixture-of-experts model engineered for resource-constrained private deployments that need strong reasoning, code, and multilingual capability without the footprint of a dense 70B+ system.
gpt-oss-120b-GGUF
120B MoE reasoning model for private deployment—handle complex agentic workflows and R&D tasks on your own hardware without API dependencies.
QwQ-32B
A 32B reasoning model for private deployment—built to solve hard operational problems (support escalation, complex process automation, knowledge synthesis) where chain-of-thought depth beats speed.
granite-3.0-8b-instruct
A 8B instruction-tuned dense transformer for building private, multilingual AI assistants and automating operational workflows without leaving your infrastructure.
Qwen3-Coder-30B-A3B-Instruct-MLX-8bit
Apple Silicon-optimized coding LLM for private, on-device ops automation and custom AI applications requiring code generation and reasoning.
Qwen3-Coder-30B-A3B-Instruct-MLX-5bit
5-bit quantized coding LLM optimized for Apple Silicon—run a capable code-generation engine entirely on-device for private ops automation.
Qwen3-Coder-30B-A3B-Instruct-MLX-6bit
6-bit quantized code-generation model optimized for Apple Silicon, designed for on-device private deployment in ops workflows requiring code generation, documentation automation, and developer-facing AI.
llama-68m
Ultra-lightweight speculative decoder for private LLM inference acceleration and edge deployment in ops workflows.
gemma-4-26B-A4B-it-uncensored
Uncensored 26B MoE model for companies building internal AI agents and operational automation where refusal behavior is a workflow blocker.
OLMoE-1B-7B-0924
A 1B-active-parameter Mixture-of-Experts model optimized for cost-efficient private deployment in ops automation and custom AI workflows where performance-per-token matters.
Step-3.7-Flash-NVFP4
Production-grade sparse MoE vision-language model for high-throughput agentic automation, financial document processing, and multi-step operational workflows in private environments.
codegen-350M-mono
A lightweight Python code-generation model for automating developer workflows and embedding code synthesis into private operational systems.
Qwen3.6-27B-GGUF
Dense 27B coding engine for on-device agentic AI—where you need all parameters active per token for long-horizon tool chains, repo-level reasoning, and full data control.
granite-4.0-tiny-preview
A 7B MoE instruct model for private-hosted ops workflows—reasoning, summarization, doc handling, and function-calling in your own infrastructure.
Qwen3-Coder-Next-AWQ-4bit
MoE coding agent for private deployment—3B active params, 256k context, tool-calling backbone for agentic ops automation.
DeepSeek-V4-Pro-NVFP4
A 1.6T-parameter MoE model quantized to 4-bit for private deployment—built for reasoning-heavy ops automation, agentic workflows, and custom AI on your own infrastructure.
gemma-4-E4B-it-OBLITERATED
A 7.9B parameter base model with guardrails surgically removed — built for operators who need an unrestricted private LLM they fully control.
Qwen3-4B-Instruct-2507-NVFP4
A 4B parameter quantized instruction-tuned model optimized for self-hosted inference on constrained hardware, enabling private operational AI deployments without cloud dependencies.
Qwen2.5-Coder-7B-Instruct-GGUF
A 7B code-specialized LLM in GGUF format, built for self-hosted code generation, reasoning, and fixing within private operational environments.
Qwen2.5-32B-Instruct-fp8-dynamic
A 32B instruction-tuned model optimized for coding, math, long-context reasoning, and structured output—deployable privately to automate knowledge work and build custom AI agents without external API dependency.
Qwythos-9B-Claude-Mythos-5-1M
A 9B reasoning model with 1M context and native tool-use—built for private deployment in ops automation, agentic workflows, and custom AI systems that need to stay in your infrastructure.
LightOnOCR-2-1B
A 1B vision-language model for document OCR and text extraction that runs efficiently on modest hardware—purpose-built for companies automating document processing at scale within their own infrastructure.
Qwen3-8B-NVFP4
FP4-quantized 8B base model optimized for private inference on NVIDIA GPUs—purpose-built for ops teams deploying custom agents, RAG systems, and internal automation without cloud vendor lock-in.
gpt-oss-safeguard-20b
Purpose-built safety classifier for private LLM pipelines: filter inputs/outputs, label content, and enforce custom policies without leaving your infrastructure.
granite-4.1-3b-GGUF
Compact 3B base model in GGUF format for private, resource-constrained deployments of custom operational AI and internal automation.
Qwen2.5-Coder-14B-Instruct-MLX-4bit
Apache-2.0 code-generation model optimized for private deployment on Apple Silicon, designed to automate internal code tasks and power custom AI agents within ops teams.
Qwen2.5-0.5B-Instruct-GGUF
Ultra-lightweight instruction-tuned LLM for private deployment in resource-constrained ops environments—0.5B parameters, GGUF-quantized, runs on CPU/edge hardware.
sundial-base-128m
Purpose-built time-series foundation model for private forecasting automation—embed zero-shot predictions into ops workflows without external APIs.
gpt2-xl
A 1.5B parameter base language model for private text generation—suitable for ops automation, document drafting, and custom AI applications where you control the data and infrastructure.
Step-3.7-Flash
High-throughput vision-language agent backbone for private, production agentic workflows—parse documents, orchestrate tools, verify cross-source data without cloud lock-in.
TinyLlama-1.1B-Chat-v0.3-AWQ
Ultra-lightweight 1.1B chat model optimized for on-device and edge inference—purpose-built for companies running private, low-latency AI operations on constrained hardware.
Qwen3-1.7B-GPTQ-Int8
A 1.7B quantized reasoning model for private ops automation—reasoning + speed in a deployable footprint.
LLaDA2.0-mini
MoE diffusion LLM for private deployment: 16B total params, 1.4B active, Apache-licensed—built to run cost-efficiently in your own infrastructure while supporting tool-use and complex reasoning.
SmolLM2-1.7B-Instruct
Lightweight instruction-tuned LLM for on-device ops automation and private business AI without GPU requirements.
GLM-4.5
A 355B MoE reasoning model (32B active) engineered for agentic workflows, tool use, and hybrid reasoning—built for companies deploying private LLM systems that require complex task automation and controlled inference.
Olmo-3-7B-Instruct
A 7B instruction-tuned model optimized for reasoning and tool use, deployable on modest hardware for private ops automation and custom AI applications.
llava-onevision-qwen2-7b-ov
Multimodal foundation for private document/image understanding and video analysis in operational workflows—built on Qwen2, Apache 2.0, runs offline.
gemma-4-31B-it-qat-q4_0-unquantized-assistant
A 31B multimodal reasoner for private, custom ops AI—text/image processing, agent workflows, and long-context automation that stays inside your infrastructure.
pythia-12b
Research-grade 12B causal LM for building interpretable, fine-tunable text-generation systems that stay entirely in your infrastructure.
Qwen2.5-Coder-32B-Instruct-GPTQ-Int4
Production-grade code LLM for private deployment: automate engineering workflows, build internal code agents, and maintain full data control in your environment.
Midm-2.0-Mini-Instruct
Korea-centric 2.3B instruct model for private, on-device deployment in ops workflows requiring localized reasoning and low-latency inference.
Qwen2.5-7B-Instruct
A 7B instruction-tuned model built for private deployment, strong at long-context reasoning and multilingual ops tasks — designed for self-hosted enterprise automation.
Mistral-7B-Instruct-v0.3-GGUF
A quantized, self-hostable 7B instruction-tuned model optimized for on-premise deployment and custom ops automation without cloud vendor lock-in.
TinyLlama-1.1B-Chat-v1.0-GPTQ
Micro-language model (1.1B params) quantized for CPU/GPU inference—designed for resource-constrained private deployments where data residency and operational automation matter more than frontier capability.
Ornith-1.0-9B
A 9B coding-agent model for companies building internal agentic automation and self-hosted DevOps workflows that need tight control over data and inference.
Ovis2.5-9B
Native-resolution multimodal LLM for enterprise document automation, chart analysis, and visual reasoning—deployable privately to keep proprietary data in-house.
SmolLM3-3B-Base
A 3B parameter base model designed for private, on-device deployment in resource-constrained ops environments where reasoning, multilingual support, and long-context handling matter.
Qwen3-4B-GGUF
Compact 4B reasoning model for private ops automation—switching on demand between deep thinking (math, logic, code) and fast inference (support, routing, summaries).
granite-3.1-8b-instruct
8B instruction-tuned model designed for enterprise ops automation, RAG, document processing, and custom AI agents—sized for cost-effective self-hosted deployment.
pythia-14m
Research-grade 14M-parameter base model for building interpretable, lightweight custom AI systems and operational automations that run entirely on-premises.
Qwen3-235B-A22B-FP8
A 235B mixture-of-experts model with native thinking/non-thinking modes—designed for companies building private reasoning agents, complex workflow automation, and multilingual ops AI without external API dependencies.
Qwen3-30B-A3B-GPTQ-Int4
A 30B MoE model with switchable thinking/non-thinking modes—built for private deployment to automate complex reasoning tasks while keeping data on-premise.
Mixtral-8x22B-v0.1-GGUF
A 176B mixture-of-experts model in GGUF quantized format: designed for companies deploying large-scale language reasoning on-premises with tunable memory trade-offs.
biogpt
Domain-specific generative model for biomedical text generation and relation extraction — built to automate documentation, literature synthesis, and clinical knowledge tasks within regulated environments.
OLMoE-1B-7B-0125-Instruct
Lightweight mixture-of-experts model for private ops automation and custom chat/reasoning applications where you control the compute and data.
VertaLily-1.2-1B-GGUF
Lightweight 1B private LLM for edge deployment, local workflows, and ops automation where data must stay in-house.
Qwen2.5-7B-Instruct-GPTQ-Int4
Compact 7B instruction-tuned model optimized for private deployment and operational automation—4-bit quantized for resource-constrained environments without sacrificing coding, math, and long-context capability.
pythia-160m-deduped
Small, interpretability-focused base model for research and fine-tuning; suitable for private deployment in resource-constrained ops environments.
Ovis2-1B
Lightweight multimodal LLM for private document/image understanding and operational automation at 1.3B parameters.
ctrl
Conditional text generation model with explicit control codes—designed for domain-specific, templated content automation in ops workflows where you need predictable, steerable outputs.
Qwen3.5-122B-A10B-NVFP4
Quantized 122B MoE model optimized for private, GPU-accelerated inference in ops workflows, RAG, and agent systems—keeping data in-house.
EAGLE-LLaMA3-Instruct-8B
Speculative decoding acceleration layer for private LLM deployments—cut inference latency 3–5.6x without retraining your base model.
Ornith-1.0-35B-FP8
Specialized agentic coding model for private deployment—automates software engineering workflows (repo navigation, bug fixing, code generation) entirely within your environment.
lynx-instruct-30b
A Qwen3-based MoE model tuned for Nordic/European languages, deployable self-hosted to automate multilingual ops workflows while keeping data on-premise.
Qwen2.5-Coder-14B-Instruct-MLX-8bit
Apple Silicon–optimized code LLM for private, agent-driven automation of engineering and documentation workflows.
Qwen2.5-32B-Instruct-GPTQ-Int8
32B instruction-tuned model optimized for private deployment via GPTQ 8-bit quantization—built for ops teams running custom AI without cloud dependency.
Qwen3-Coder-Next-AWQ-4bit
A 14B-parameter coding agent model (3B active via MoE) designed for private, tool-calling automation in engineering ops and code-heavy workflows.
MiniCPM4.1-8B
8B sparse-attention reasoning model built for private deployment, enabling companies to automate complex ops workflows (support triage, document analysis, decision automation) while keeping all customer/internal data on-premises.
Qwen2.5-7B-Instruct-unsloth-bnb-4bit
A quantized 7B instruction-tuned model optimized for cost-effective private deployment and rapid fine-tuning in resource-constrained ops environments.
Qwen3-8B-DFlash-b16
A lightweight diffusion-based speculative drafter that accelerates Qwen3-8B inference in private deployments by 6x+ via parallel token prediction.
Mistral-Small-24B-Instruct-2501-GGUF
24B instruction-tuned model in GGUF format: designed for efficient private deployment and custom ops automation without vendor lock-in.
Qwen2.5-Math-7B
Specialized math reasoning engine for private deployment—solves English/Chinese math problems via chain-of-thought and tool-integrated reasoning, suitable for ops teams automating quantitative workflows.
SmolLM2-1.7B
A 1.7B lightweight language model optimized for on-device and self-hosted deployment in operational workflows—instruction-following, reasoning, and code tasks without the infrastructure cost of larger models.
Qwen2.5-14B-Instruct-GPTQ-Int4
14B instruction-tuned model with 4-bit quantization for private, cost-efficient deployment in ops workflows—coding, long-context document processing, and structured data handling.
WizardLM-2-7B-GGUF
Compact 7B instruction-tuned model for private ops automation and custom AI—fast enough for edge/self-hosted, capable enough for reasoning, coding, and multi-turn workflows.
Yi-1.5-6B-Chat-GGUF
A 6B quantized chat model for private, on-device operational automation—small enough for edge/desktop deployment, strong enough for internal support and workflow tasks.
Qwen3-8B-GGUF
8B reasoning model with switchable thinking/non-thinking modes—designed for ops teams building private agents, internal automations, and cost-efficient custom AI without cloud inference.
Olmo-3-1025-7B
A 7B base model for building proprietary AI workflows and autonomous agents that run entirely in your infrastructure.
Nex-N2-mini-AWQ-INT4
Agentic reasoning model optimized for long-horizon task automation, tool use, and code execution—built for ops teams running multi-step workflows on private infrastructure.
xglm-564M
Lightweight multilingual text generation for private ops automation across 30 languages—no external API, data stays in-house.
Qwen3-Coder-Next-AWQ-8bit
A 3B-activated, 80B-parameter MoE coding model designed for private, agent-driven development automation and tool-calling workflows running entirely on customer infrastructure.
Hy-MT2-30B-A3B
Specialized multilingual translation engine for private deployment—33-language support with instruction-following and style control, built for companies automating global content and documentation workflows.
BioMistral-7B
Domain-specialized 7B LLM for medical/biomedical operational workflows—QA systems, clinical documentation analysis, knowledge extraction—deployable entirely on-premise with no data leaving your infrastructure.
Ornith-1.0-397B-FP8
Ornith-1.0-397B is a self-hosted, MIT-licensed coding-agent model for enterprises building private AI systems that automate software engineering and agentic task execution without sending code or data to external APIs.
LLaDA-8B-Base
8B diffusion-based LLM for private deployment in ops workflows—text generation, automation, and custom AI without external API dependency.
Apertus-8B-Instruct-2509
An 8B multilingual instruction model built entirely on open data and compliant training practices—designed for companies running private AI without legal liability or data leakage risk.
DeepSeek-V3-0324-GGUF
Quantized, locally-runnable inference engine for DeepSeek-V3—built to run reasoning-heavy workloads on modest hardware without cloud dependencies.
tinyllama-bnb-4bit
Lightweight 1.1B quantized LLM for resource-constrained private deployments, fine-tuning automation, and operational task automation on edge/on-prem infrastructure.
Qwen3-Next-80B-A3B-Thinking-AWQ-4bit
80B sparse reasoning model for private deployment—built for ops teams automating complex workflows and custom AI systems that need to stay behind your firewall.
MiniCPM3-4B
Lightweight 4B instruction-tuned model for private deployment and ops automation in cost-sensitive, resource-constrained environments.
Qwen3-Coder-Next-NVFP4
Production-grade code-reasoning backbone for private AI ops—80B MoE model compressed to 45GB, built to run on customer infrastructure with minimal latency.
mxbai-rerank-base-v2
A multilingual reranker for retrieval pipelines—embed search results, customer queries, or knowledge-base hits to rank relevance without replacing your primary retriever.
typhoon2.5-qwen3-4b
A 4B Thai/English instruction-tuned model with 256K context and function-calling—built for ops teams automating multilingual workflows and deploying private AI agents that stay within your infrastructure.
gpt-neox-japanese-2.7b
A 2.7B Japanese LLM optimized for private deployment in Japanese-speaking enterprises automating internal documentation, customer support, and knowledge workflows.
MiMo-V2.5-Pro
1T-parameter MoE model engineered for long-context agentic workflows and complex ops automation—42B active params, 1M token window, FP8 native.
GLM-5.1
Agentic coding and operational task automation—an open 753B MoE model engineered for long-horizon problem-solving, tool use, and iterative refinement in private deployments.
Qwen3.6-27B-MTP-pi-tune-GGUF
A 27B dense LLM fine-tuned for fast agentic loops—direct tool calls, no reasoning overhead—packaged as GGUF for private, single-machine deployment.
Qwen3-0.6B-GGUF
Ultra-lightweight reasoning model (0.6B) for private, CPU-friendly ops automation—thinking mode for logic-heavy tasks, fast mode for high-volume conversational workflows.
Qwen3-8B
8B reasoning model with switchable thinking/non-thinking modes—built for private ops automation, custom agent workflows, and cost-sensitive reasoning tasks.
Qwen3-32B-FP8
A 32B reasoning model with switchable thinking/non-thinking modes, built for private deployment in ops workflows requiring both complex reasoning and fast inference.
Qwen3-VL-30B-A3B-Thinking-AWQ
Vision-language agent for automating visual workflow tasks, document intelligence, and GUI-based operations in private environments—keep all image/video/screen data inside your infrastructure.
Qwen2.5-7B-Instruct-GGUF
A compact 7B instruction-tuned model designed for private, on-premise deployment—enabling ops teams to build custom AI workflows without cloud dependency or data leakage.
Qwen2.5-7B-Instruct-GGUF
Lightweight 7B instruction-tuned model in GGUF format—deploy on modest CPU/GPU hardware for private, cost-controlled conversational AI and ops automation.
MiniMax-M2.7-REAP-172B-A10B-NVFP4-GB10
Dense-expert MoE (172B) quantized for private inference on enterprise hardware—trade-off model for ops teams needing reasoning + code in a self-hosted, cost-controlled footprint.
Qwen2.5-Coder-7B-Instruct-bnb-4bit
Apache-2.0 code LLM (7B, 4-bit quantized) for self-hosted code automation, agent logic, and custom development workflows—built to run lean on customer infrastructure.
Hy-MT2-1.8B
Specialized multilingual translation engine for private deployment—33-language support with instruction-following for enterprise document workflows, customer support localization, and real-time content translation without cloud dependency.
granite-3.3-8b-instruct
An 8B instruction-tuned model built for private deployment into ops workflows—reasoning, code, extraction, RAG—where data stays in your environment and you control the inference.
gpt-oss-20b-BF16
Apache 2.0 open-weight MoE model (21B params, 3.6B active) purpose-built for on-premise reasoning, agentic automation, and fine-tuning on consumer/enterprise hardware.
stablelm-3b-4e1t
A 3B decoder-only base model for fine-tuning into private, task-specific AI agents and operational automation within your own infrastructure.
Phi-3.5-mini-instruct-GGUF
Lightweight instruction-tuned model in GGUF format for CPU/edge inference—minimal ops footprint, maximum data control.
Ornith-1.0-397B
Agentic coding backbone for private deployments: 397B MoE model purpose-built to automate software engineering workflows and terminal tasks without sending code to external APIs.
GLM-4.7-Flash-GGUF
Lightweight 30B MoE model built for on-premise ops automation, tool-calling, and custom AI without external API dependency.
Qwen3-8B-GGUF
8B dense model with native thinking/non-thinking modes, built for private deployment in ops workflows requiring reasoning agility without enterprise LLM costs.
Qwen3-30B-A3B-Base
A 30B parameter mixture-of-experts model for private deployment in ops workflows—reasoning, coding, multilingual support—with only 3.3B active parameters per inference pass, minimizing hardware and latency.
Qwen2.5-Coder-32B-Instruct-GGUF
32B code-specialized LLM optimized for private deployment via GGUF quantization—purpose-built for enterprises automating software development workflows, code review, and technical documentation without external API dependency.
LLaDA2.1-flash
A 100B diffusion-based LLM optimized for speed/quality trade-offs in private deployments, enabling ops teams to automate knowledge work without latency penalties.
Qwen2.5-Coder-0.5B-Instruct
A 0.5B code-specialized instruction model for embedding code generation, debugging, and agent workflows into private operational systems without the footprint of larger codeLLMs.
sarvam-30b-FP8-dynamic
FP8-quantized 30B multilingual model optimized for private inference—purpose-built for ops teams running proprietary workloads on controlled hardware.
gemma-4-E2B-it-assistant
Lightweight multimodal reasoning engine for on-device ops automation, document processing, and private agentic workflows—built for deployment where data never leaves your infrastructure.
SmolLM2-360M
Lightweight edge-deployable base model for custom ops automation, on-device private inference, and internal workflow agents where data residency and model control are non-negotiable.
Ovis1.6-Llama3.2-3B
Edge-optimized multimodal LLM (3B) for private, on-device image-text reasoning in operational workflows—vision tasks that must stay in your environment.
GLM-5.2-NVFP4
744B sparse MoE model (40B active params) for ops teams building private, cost-controlled AI agents and custom applications without vendor lock-in.
RaDialog-interactive-radiology-report-generation
Vision-language model purpose-built for automated radiology report generation and radiologist-assistant dialogue—run privately in healthcare ops to keep imaging data in-house.
Qwen3.5-122B-A10B-NVFP4
A 122B MoE model quantized to 4-bit for cost-effective private inference on NVIDIA hardware—purpose-built for ops teams automating multi-turn agent workflows, RAG systems, and departmental automation without shipping data to third parties.
Qwen3-8B-speculator.eagle3
A speculative decoding accelerator for Qwen3-8B that trades inference latency for throughput—designed for ops teams running private, cost-conscious deployments at scale.
Ovis1.6-Gemma2-9B
Vision-language model (10B) for automating document analysis, image-based workflows, and visual reasoning tasks in self-hosted ops environments.
Qwen3Guard-Gen-8B
A specialized safety-classification model for building content moderation pipelines into custom AI applications—designed to run entirely within your infrastructure.
3b-de-ft-research_release
A 3.3B German-language LLM fine-tuned for text generation, deployable privately to automate German-language ops workflows without cloud dependencies.
Qwen2.5-14B-bnb-4bit
A 14B quantized base model for ops teams building private, instruction-tuned agents and automating internal workflows without shipping data to external APIs.
Gemma-4-26B-A4B-it-NVFP4
Production-ready quantized MoE model for private, cost-efficient inference on modern hardware—deploy multi-billion-parameter reasoning in your own environment.
GDN-primed-HQwen3-8B-Instruct
Long-context hybrid inference engine for ops teams building private AI systems that need 2× throughput gains on document processing, knowledge retrieval, and agent workflows without sacrificing model quality.
Qwen3-235B-A22B-Instruct-2507
235B sparse mixture-of-experts model for private, long-context agentic automation and custom knowledge-intensive ops AI.
Mistral-7B-Instruct-v0.2-GGUF
Quantized 7B instruction-tuned model for CPU-first private deployment and operational automation workflows.
Laguna-XS.2
A 33B MoE coding agent that runs locally on commodity hardware—purpose-built for companies automating technical workflows and building private AI systems that keep code, logs, and operational data on-premises.
gte-Qwen2-7B-instruct
Production-grade multilingual text embedding model for retrieval, semantic search, and knowledge-graph indexing in private, self-hosted deployments.
GLM-5.1-NVFP4
Quantized 754B MoE model (40B active) for private deployment of agentic AI systems, RAG, and operational chatbots on NVIDIA infrastructure—data stays in your environment.
DeepSeek-R1-0528-Qwen3-8B-GGUF
A 8B chain-of-thought model optimized for reasoning-heavy operational tasks, deployable fully private on modest GPU/CPU hardware.
internlm3-8b-instruct
8B reasoning model for private-hosted ops automation—reasoning, knowledge work, and custom workflows with data control.
LongCat-Flash-Chat
560B MoE model optimized for agentic workflows and operational task automation with dynamic parameter activation (~27B active) for cost-effective private deployment.
gemma-4-12B-it-assistant
A multimodal reasoning model (12B dense) designed for private deployment across ops workflows—document processing, agentic automation, and long-context tasks on enterprise infrastructure.
DeepSeek-V4-Flash-DSpark
Lightweight MoE model for private deployment in ops workflows: 13B activated parameters, 1M token context, MIT-licensed for unrestricted self-hosting.
Qwen3-4B-unsloth-bnb-4bit
4B reasoning model for private ops automation: thinking + non-thinking modes in a single deployable engine.
Qwen2.5-14B-Instruct-GGUF
A quantized 14B instruction-tuned model for private deployment in resource-constrained ops environments, trading minimal inference cost against quality.
VibeThinker-3B
A 3B reasoning specialist for private deployment—math, code, STEM verification tasks where answer correctness is measurable and data stays in your environment.
Qwen3-4B-FP8
A 4B parameter dense model with switchable thinking/non-thinking modes, built for companies deploying lightweight reasoning agents and operational automation in private environments.
Qwen2.5-1.5B-Instruct
Edge-optimized 1.5B instruction-tuned model for private, on-device AI automation—runs locally on Android/iOS with <2GB memory.
VibeThinker-3B-Q4_K_M-GGUF
Lightweight 3B reasoning model in GGUF format for on-device AI agents and private ops automation on consumer/mid-range GPUs.
granite-4.1-30b
Apache-2.0 30B instruction-tuned model purpose-built for tool-calling, multi-step reasoning, and agentic workflows in private enterprise deployments.
Qwen3-30B-A3B-NVFP4
FP4-quantized 30B MoE model optimized for private inference on NVIDIA hardware—fast, memory-efficient ops AI and agent deployment without cloud dependency.
TinyLlama-1.1B-intermediate-step-1431k-3T
1.1B Llama-compatible model for companies building lightweight private AI agents, internal automation, and cost-controlled custom applications without cloud dependency.
DeepSeek-R1-Distill-Qwen-1.5B
Edge-optimized 1.5B distilled reasoner for on-device private ops automation and customer-facing AI on Android/iOS.
Qwen3-0.6B-unsloth-bnb-4bit
Lightweight reasoning model (0.6B) for private, on-device operational AI and custom applications requiring thinking-mode logic without GPU overhead.
svara-tts-v1
Private-deployable multilingual TTS for Indic languages—runs on commodity hardware, keeps audio generation in your environment, supports emotion/style control and LoRA customization for domain-specific voices.
DeepSeek-R1-Distill-Qwen-1.5B-GGUF
Lightweight reasoning model (1.5B) for private deployment in ops workflows—fast inference, chain-of-thought reasoning, runs on modest CPU/GPU.
Qwen3-14B-NVFP4
FP4-quantized 14B base model optimized for private inference on NVIDIA GPUs—drop-in for RAG, agents, and operational automation without external API calls.
MiniCPM4-0.5B
Ultra-lightweight LLM (0.5B) for edge-side deployment and private ops automation on resource-constrained infrastructure.
DeepSeek-R1-0528-Qwen3-8B-GGUF
A compact 8B reasoning model (distilled from DeepSeek-R1-0528) designed for private deployment in ops workflows requiring math, code, and logic reasoning without external API dependency.
pythia-1.4b
A 1.4B interpretability-research baseline for building controlled, auditable custom AI agents and operational automation on private infrastructure.
Qwen3-4B-MLX-4bit
A 4B parameter reasoning model designed for self-hosted operational AI: instruction-following, tool calling, and multi-language workflows that run entirely on customer infrastructure.
GLM-4.7-AWQ
A 358B mixture-of-experts coding and reasoning model optimized for private deployment via AWQ quantization, designed for ops teams building custom AI agents that control code generation, terminal tasks, and complex multi-turn workflows.
GLM-4.7-FP8
A 358B open-weight MoE model optimized for coding, reasoning, and agentic automation—designed for companies building private AI workflows and operational automation systems.
Qwen2.5-Coder-14B-Instruct-GGUF
Code-specialized 14B instruction-tuned model in GGUF format, built for private deployment in custom AI agents and operational automation workflows.
OLMo-2-0425-1B-Instruct
A 1.3B instruction-tuned model designed for self-hosted operational AI and custom applications where data residency and model control matter.
Kimi-Linear-48B-A3B-Instruct
A 48B linear-attention model engineered for long-context operational tasks and private deployment—trades full-attention precision for 6× faster decoding and 75% KV-cache reduction.
granite-3.0-1b-a400m-base
Lightweight sparse MoE for private, cost-controlled text generation in ops workflows—summarization, classification, extraction, QA—where inference speed and VRAM footprint matter.
Ouro-1.4B
A 1.4B parameter looped language model designed for parameter-efficient reasoning in private deployments, trading recurrent compute for model size to run cost-effectively on edge/internal infrastructure.
Qwen2-7B
Base model for building private, custom AI systems that need strong reasoning, coding, and multilingual understanding without vendor lock-in.
Qwen2.5-3B-Instruct-bnb-4bit
3B instruction-tuned model optimized for fast, resource-efficient private deployment in operational AI systems—coding, math, structured data, and long-context tasks.
granite-docling-258M
Document-to-structured-data converter: automate PDF/image ingestion, layout parsing, and content extraction for internal document workflows running entirely on your infrastructure.
BMOJOF-primed-HQwen3-8B-Instruct
A hybrid Attention+SSM model engineered for long-context ops workflows at 1.75× faster inference than standard transformers—designed to run private, self-hosted, with reduced memory footprint for high-throughput batch processing.
tiny-random-qwen3
A minimal debugging/test version of Qwen3-4B for validating LLaMA Factory workflows and private model deployment patterns without production overhead.
stories15M_MOE
A 36M-parameter mixture-of-experts model for testing narrative generation; suitable only for private prototyping, not production ops automation.
OpenReasoning-Nemotron-32B
Post-trained reasoning engine for math, code, and science—built to run privately and scaled via ensemble inference for competitive accuracy without external APIs.
gpt-oss-20b-unsloth-bnb-4bit
Apache 2.0 open-weight MoE reasoning model (20B params, 3.6B active) for self-hosted ops automation, agentic workflows, and fine-tuned custom AI—fits 16GB VRAM.
Qwen3-Next-80B-A3B-Instruct-AWQ-4bit
80B sparse MoE model for private deployment at scale — use it to automate complex ops workflows, reasoning tasks, and long-context agent applications while keeping data in your own environment.
Phi-4-mini-instruct-GGUF
A lightweight 3.8B reasoning-focused model for private deployment in compute-constrained ops environments; designed for custom task automation, internal agent workflows, and data-local inference.
ERNIE-4.5-21B-A3B-PT
MoE text model (21B total, 3B active) for private deployment in ops automation, document processing, and internal knowledge workflows where data residency and cost efficiency matter.
MiMo-V2-Flash
MiMo-V2-Flash: a 309B MoE model with 15B active parameters—designed for high-throughput private inference, agentic workflows, and ops automation where inference cost and latency matter.
Tongyi-DeepResearch-30B-A3B
A 30B MoE agent model designed for research automation and long-horizon agentic workflows—deploy privately to control data while automating complex information-seeking tasks.
Fanar-1-9B-Instruct
Arabic-English 9B model for building private, culturally-aware conversational AI and ops automation in environments where data residency and dialect support matter.
Qwen2.5-Coder-14B-Instruct-GGUF
Code-specific 14B instruction-tuned model in GGUF format—lightweight enough to self-host, capable enough to power code agents and internal dev automation on private infrastructure.
Huihui-Qwythos-9B-Claude-Mythos-5-1M-abliterated-GGUF
A 9B abliterated reasoning model in GGUF format for companies building private, uncensored custom AI agents and automating sensitive operational workflows.
GLM-5
744B MoE model for agentic reasoning, code, and complex operational automation—built for private deployment and tool-use workflows.
tiny-random-gpt-oss-mxfp4
A tiny, quantized text-generation model for rapid prototyping of conversational AI in resource-constrained private environments.
OpenThinker3-7B
A 7B reasoning model fine-tuned for math, code, and science tasks—designed for ops teams building private, custom reasoning agents that stay on-premise.
Qwen2.5-14B-Instruct-4bit
A 14B instruction-tuned model optimized for MLX (Apple Silicon) deployment—designed for companies running private, conversational AI workloads on their own hardware without cloud dependency.
Qwen2.5-7B-Instruct-GPTQ-Int8
A 7B instruction-tuned model optimized for private deployment via 8-bit GPTQ quantization, enabling ops teams to run high-quality reasoning, coding, and structured output tasks on modest hardware without external API calls.
OLMo-1B-hf
1B base model for companies building lightweight private AI agents and automating operational workflows without cloud dependency.
Mamba2-primed-HQwen3-8B-Instruct
A hybrid attention-SSM 8B model for ops teams deploying long-context AI at 2× inference speed in their own infrastructure, trading some reasoning quality for throughput and memory efficiency.
Qwen3.6-27B-DFlash
Speculative decoding drafter component for accelerating Qwen3.6-27B inference in private, self-hosted deployments—trades a lightweight auxiliary model for 2-3× throughput gains in batch/online serving.
Qwen3-4B-Instruct-2507-unsloth-bnb-4bit
4B instruction-tuned model optimized for agentic ops automation and private deployment—strong reasoning, tool-calling, and 256K context for internal knowledge workflows.
Qwen2.5-Coder-7B-Instruct-AWQ
A 7B code-focused instruction-tuned model optimized for private deployment in engineering ops, internal code agents, and custom AI workflows where data residency matters.
tiny-random-PhiForCausalLM
Tiny reference/testing model for validating Phi-architecture pipelines in private, resource-constrained environments—not production.
Qwen2.5-14B
A 14.7B foundational model for companies building private, domain-specific AI agents and automating structured operational workflows—coding, math, JSON generation, and long-context reasoning without vendor lock-in.
GLM-4-32B-0414.w4a16-gptq
A 4-bit quantized 32B reasoning model designed for self-hosted private deployment on consumer/mid-market hardware, optimized for operational automation and internal knowledge workflows.
GLM-4.7-Flash-FP8-Dynamic
30B MoE model optimized for private deployment and operational automation—reasoning, coding, and agentic tasks without external API dependencies.
Qwen-AgentWorld-35B-A3B
A native language world model for simulating agentic environments across 7 operational domains—tool use, search, terminals, software engineering, mobile, web, and OS—enabling companies to build private AI agents that understand and predict system state changes.
gemma-4-12b-heretic-abliterated-GGUF
A lean, fully quantized 12B generalist for private agentic workflows and custom AI automation where you control the model and data completely.
SmolLM-1.7B
Compact, self-contained language model for private deployment in ops workflows—code, document automation, and internal knowledge tasks where data stays on-premise.
Qwen2.5-1.5B-Instruct-unsloth-bnb-4bit
Ultra-lightweight instruction-tuned LLM (1.5B) optimized for on-device private deployment and fine-tuning in resource-constrained ops environments.
GLM-5.2-Int4-Int8Mix
A 785B open-weight MoE model with 1M context and mixed-precision quantization (INT4/INT8) designed for private deployment on multi-GPU infrastructure, balancing reasoning depth with inference cost in ops-critical workflows.
Jan-nano-128k
Compact 4B model with native 128k context for document-heavy research automation and long-form operational workflows that run entirely in your infrastructure.
phi-1_5
Lightweight code and reasoning engine for private ops automation—QA, code generation, and document summarization without leaving your infrastructure.
gemma-4-12B-it-Claude-4.6-4.8-Opus-GGUF
A 12B reasoning model in GGUF format, tuned on Claude reasoning data, designed to run fully private on modest hardware (4.5 GB VRAM minimum) for internal ops automation and custom AI without external API dependencies.
GLM-5.1-GGUF
Agentic coding & ops AI base model designed for long-horizon tool use, built to run privately and fine-tune for custom automation workflows.
TinyLLama-v0
Ultra-lightweight Llama-architecture model for private, resource-constrained ops automation and custom story/narrative generation tasks.
gpt2-mini
Lightweight text-generation baseline for private, resource-constrained deployments where a company needs full data control and fast inference on modest hardware.
sarvam-105b
A 105B-parameter MoE model optimized for reasoning, agentic tasks, and 22 Indian languages—designed for enterprises building private AI systems that require strong logic and multi-lingual operational intelligence without vendor lock-in.
VibeThinker-3B-GGUF
A compact 3B reasoning engine for private deployment—math, coding, and STEM verification tasks where you own the inference stack and control every token.
Qwen2.5-Coder-14B-Instruct-GPTQ-Int8
A 14B code-specialist LLM optimized for private code generation, agent automation, and internal developer tooling—quantized for modest hardware footprints.
lumeleto
A compact, fine-tuned text-generation model for ops teams building private AI workflows on cost-constrained hardware.
Qwen3-14B-Base
A 14.8B dense foundation model for private deployment in ops workflows—reasoning, multilingual processing, and custom AI agents without vendor lock-in.
Qwen3-4B-Instruct-2507-MLX-4bit
Lightweight 4B instruction-tuned model optimized for Apple Silicon; purpose-built for private deployment in ops workflows where data residency and inference speed matter more than frontier capability.
Qwen3-8B-FP8-dynamic
FP8-quantized 8B reasoning model optimized for cost-efficient private deployment on single-GPU infrastructure, validated on Red Hat platforms.
Qwen2-1.5B-Instruct-AWQ
A 1.5B instruction-tuned model optimized for private, on-premise deployment to automate conversational operational tasks without cloud dependencies.
sarvam-30b
A 30B MoE model optimized for private deployment in resource-constrained ops environments, with strong multilingual (22 Indian languages) and tool-use capability for enterprise automation.
Dream-v0-Instruct-7B
A 7B instruct-tuned open-weight model for private deployment in ops workflows—document processing, support automation, and knowledge work without external API dependency.
Ornith-1.0-9B-MTP-GGUF
9B inference engine optimized for self-hosted private deployment with speculative decoding—run fast, quantized, on your own hardware.
OLMo-7B
A 7B open-weight base model for companies building private language systems, fine-tuned agents, and internal automation—trained on transparent data (Dolma) with full inference/training code available.
Qwen3.6-14B-A3B-FableVibes-GGUF
A pruned 14B MoE reasoning model distilled from Claude Fable traces—runs on consumer hardware while preserving structured multi-step reasoning for internal ops automation and custom AI applications.
Qwen3-Embedding-4B-AWQ-INT4
Lightweight embedding model (4B, INT4-quantized) designed to run privately on consumer GPUs for semantic search, retrieval, and custom knowledge applications without cloud dependencies.
Qwen2.5-Coder-1.5B-Instruct-GGUF
1.5B code-generation model in GGUF format—runs locally on modest hardware, purpose-built for automating code-heavy operational workflows and embedding in private AI agents without external API calls.
LLaMA-1B-dj-refine-150B
A lightweight 1.3B parameter LLM optimized for private deployment and fine-tuning, trained on curated data to punch above its weight class—ideal for ops teams building cost-effective custom AI without external API dependencies.
Huihui-Qwen3.6-35B-A3B-Claude-4.7-Opus-abliterated
Uncensored 35B reasoning model for private deployment in ops workflows where safety filtering overhead isn't acceptable; distilled from Claude 4.7 via abliteration.
Qwen3.6-35B-A3B-Abliterated-Heretic-AWQ-4bit
A 4-bit quantized, 36B multimodal MoE model for private deployment in ops automation, custom AI agents, and internal knowledge systems where data stays behind your firewall.
granite-3.1-2b-instruct-quantized.w4a16
Lightweight 2B instruction-tuned model designed for cost-effective private deployment in operational workflows—runs on commodity hardware with 75% smaller footprint than FP16.
granite-guardian-4.1-8b
A specialized safety and evaluation model for private AI systems: judges prompts, responses, and agent outputs against custom criteria without sending data to external APIs.
Qwen3-1.7B-FP8
A 1.7B quantized reasoning model for private, edge-deployable ops automation—thinking/non-thinking toggle for complex workflows without external API dependency.
OLMo-2-1124-7B-Instruct
A 7B fully open-weight instruct model designed for private deployment and custom fine-tuning in ops workflows—trained with RLVR and DPO for instruction-following across chat, reasoning, and task automation.
llm-jp-4-8b-thinking
8B dense transformer for bilingual (EN/JA) reasoning tasks — designed for private deployment in ops workflows requiring cost-efficient, locally-controlled inference.
GLM-4.7
Enterprise coding and agentic reasoning engine for private deployment—build autonomous ops workflows, internal coding agents, and complex multi-turn reasoning tasks without exposing data to third-party APIs.
Phi-4-mini-reasoning-MLX-4bit
A 600M quantized reasoning model optimized for Apple Silicon—designed for private, on-device operational AI that runs inference without cloud dependency.
Trinity-Mini-GGUF
A compact 26B MoE model (3B active) built for reasoning tasks and ops automation—runs locally via GGUF, trainable for custom enterprise workflows.
gemma-4-12B-it-qat-q4_0-unquantized-assistant
A 12B multimodal instruction-tuned model optimized for private deployment via QAT, enabling ops teams to run reasoning, coding, and document-heavy workflows entirely on-premises.
Qwen2.5-Coder-0.5B
A 0.5B code-specialized LLM for embedding in private ops workflows, agent systems, and internal code-automation tools where data stays on-premise.
Qwen3-4B-Instruct-2507-MLX-8bit
A compact 4B instruction-tuned model optimized for Apple Silicon private deployment, targeting ops teams building cost-efficient self-hosted AI agents and automation.
Qwen3-4B-Instruct-2507-MLX-5bit
Lightweight 4B instruction-tuned model optimized for Apple Silicon, quantized to 5-bit for fast private inference in resource-constrained ops environments.
Qwen3-30B-A3B-FP8-Dynamic
Quantized 30B MoE model for cost-efficient private deployment of reasoning, function-calling, and multilingual ops workflows.
Qwen3-4B-Instruct-2507-MLX-6bit
4B instruction-tuned model quantized to 6-bit for Apple Silicon—lightweight private deployment for operational workflows and internal AI agents.
llm-jp-3-150m
A compact, bilingual (JP/EN) pre-trained transformer for private ops automation and custom AI where data residency and low inference cost matter.
GritLM-7B-vllm
A 7B dual-task model (generation + embeddings) optimized for vLLM deployment, enabling private retrieval-augmented and agentic workflows in a single inference engine.
Qwen3-4B-Thinking-2507-MLX-4bit
Lightweight 4B reasoning model for Apple Silicon private deployments—ops automation, internal Q&A, and edge inference where data must stay local.
Qwen3-14B-MLX-4bit
A 14B quantized base model optimized for private, on-device deployment on Apple Silicon and edge hardware—ideal for teams building custom ops AI without cloud dependency.
mistral-7b-instruct-v0.3-bnb-4bit
Quantized 7B instruction-tuned model optimized for fast fine-tuning and private deployment on modest hardware; purpose-built for ops teams to customize and self-host without GPU overhead.
DeepSeek-V3-0324-NVFP4
Production-ready 396B quantized LLM for private deployment—90% smaller memory footprint, MIT-licensed, optimized for ops automation and custom AI on NVIDIA infrastructure.
Qwen2.5-Coder-7B-Instruct-GGUF
7B code-generation model in GGUF format: run locally on commodity hardware for private, versioned AI automation of code tasks without API dependency or data exfiltration.
Qwen2.5-1.5B-unsloth-bnb-4bit
Ultra-lightweight quantized base model for cost-efficient private inference, fine-tuning, and operational automation in resource-constrained environments.
Qwen3-4B-Thinking-2507-MLX-8bit
4B reasoning model in 8-bit quantization for Apple Silicon—compact enough for on-device ops automation, with thinking capability for logic-heavy internal workflows.
mxbai-rerank-large-v2
A reranking engine for retrieval-augmented search and knowledge retrieval—purpose-built to rank candidate documents for private, self-hosted semantic search in ops workflows.
Qwen3.5-122B-A10B-heretic-MTP-NVFP4
Quantized 122B MoE model optimized for high-throughput private inference on enterprise hardware—speculative decoding and tensor parallelism built in for ops automation at scale.
Qwen3-4B-Thinking-2507-MLX-6bit
Lightweight 4B reasoning model quantized for Apple Silicon—fit for private ops workflows, customer-facing automation, and internal knowledge work where compute is constrained.
Ornith-1.0-35B-MTP-APEX-GGUF
A self-improving agentic coding model optimized for private deployment: reasoning-heavy automation of software engineering workflows inside your infrastructure.
solar-pro-preview-instruct
A 22B instruction-tuned model engineered to run on a single 80GB GPU, delivering 70B-class reasoning for private, ops-focused AI workflows without scale penalties.
Ling-lite-1.5
MoE-based 16.8B model with 2.75B activated parameters—designed for resource-constrained private deployments that need strong reasoning, coding, and long-context performance without premium GPU infrastructure.
MiMo-V2.5-Pro-FP4-DFlash
Trillion-parameter MoE backbone for private, high-throughput agent automation—FP4-quantized experts + speculative decoding for cost-controlled agentic reasoning at scale.
Qwen2-1.5B-Instruct-FP8
Lightweight quantized chat model for cost-sensitive private deployment in ops workflows, support automation, and embedded AI agents.
HarmBench-Llama-2-13b-cls
A specialized behavioral classifier for red-teaming and safety validation—use it to automate harmful-output detection in your custom AI systems before they reach users.
Qwen3-Coder-Next-NVFP4
FP4-quantized coding model for private ops automation and custom AI agents that run on 1–2 GPUs without sacrificing benchmark performance.
Qwen3-Next-80B-A3B-Instruct-NVFP4
Production-grade 80B quantized LLM for private deployment in enterprise ops workflows, RAG systems, and custom AI agents—3.3× smaller footprint than FP16, TensorRT-LLM native.
Qwen2.5-Coder-32B-Instruct-MLX-4bit
A 32B code-specialized model optimized for Apple Silicon self-hosting, designed to automate code generation, review, and agent-driven development workflows in private environments.
neutrino-instruct
A 7B instruction-tuned model for private deployment in ops workflows—chat agents, internal Q&A, multi-turn reasoning—where data stays in your environment.
qmd-query-expansion-1.7B-gguf
A 1.7B specialized query-expansion engine for hybrid search pipelines—fine-tuned to transform raw search queries into structured lexical, vector, and hypothetical-document formats that operational teams can embed in private knowledge systems.
Qwen3.5-27B-OptiQ-4bit
A sensitivity-aware 4-bit quantized Qwen 27B optimized for private, Apple Silicon inference—built for ops teams running custom LLMs without cloud dependency.
Qwen3-235B-A22B-Thinking-2507
A 235B open-weight MoE reasoning model for private-hosted custom AI applications that demand complex reasoning, extended context, and operator control over inference.
gpt-neo-2.7B
A 2.7B parameter open-weight GPT-style model for companies building private text-generation pipelines and operational AI agents without external API dependencies.
acestep-5Hz-lm-4B
Text-to-music generation engine for companies building private, controlled audio-creation workflows into products, ops tools, or creative automation systems.
Qwen3.5-4B-Claude-4.6-Opus-Reasoning-Distilled-GGUF
A 4B reasoning model distilled from Claude Opus—lightweight enough for private deployment, structured enough for step-by-step operational automation and custom logic-heavy applications.
gpt_bigcode-santacoder
Code-completion engine for private, self-hosted development workflows—fill-in-the-middle code synthesis without cloud dependency or data leakage.
Qwen3-4B-Instruct-2507-AWQ-4bit
A 4B quantized instruction-tuned model for private deployment in ops workflows—reasoning, tool-calling, and document understanding without external APIs.
DialoGPT-small
A lightweight conversational model for building private, multi-turn dialogue agents in ops workflows without external API dependency.
Qwen3-30B-A3B-Thinking-2507-FP8
Expert reasoning engine for complex ops workflows: long-context analysis, multi-step agent tasks, and self-hosted decision automation.
llm-jp-4-32b-a3b-thinking
Japanese/English-focused MoE model for private deployment in ops workflows: multilingual reasoning and task automation without vendor lock-in.
Qwen3-14B-MLX-8bit
A 14B quantized foundation model optimized for Apple Silicon deployment, enabling ops teams to run conversational AI privately on consumer hardware without cloud dependencies.
Qwen3-235B-A22B-Thinking-2507-FP8
Reasoning-first MoE for private ops automation: deep task decomposition, 256K context, and MOE efficiency in a self-hosted package.
Moonlight-16B-A3B
A 16B MoE model (3B active) trained with Muon optimizer—optimized for private deployment and efficient custom ops automation without sacrificing reasoning quality.
Phi-3-mini-4k-instruct-gguf
Lightweight reasoning engine (3.8B) for private, cost-effective ops automation and custom AI in memory/latency-constrained environments.
Phi-3-medium-128k-instruct
A 14B parameter instruction-tuned model built for cost-constrained environments where companies need strong reasoning (code, math, logic) with 128K context—deployable entirely on-premise to keep proprietary data locked in your infrastructure.
pythia-2.8b
A 2.8B interpretability-research model for building controlled, lightweight private AI systems—not a production chatbot, but a solid base for ops automation and custom deployments where data stays in-house.
gpt-neo-1.3B
Lightweight, permissively-licensed causal language model for self-hosted text generation, prompt completion, and fine-tuning in resource-constrained private deployments.
Qwen2.5-Coder-32B-Instruct-MLX-8bit
32B code-specialized LLM quantized for Apple Silicon, deployable on-premise for private code automation, agent scaffolding, and operational AI workflows.
gpt-oss-120b-MLX-8bit
120B parameter open-weight model optimized for Apple Silicon private deployment; suitable for companies building in-house conversational AI and operational automation without cloud dependency.
rnj-1
8B dense model optimized for code, math, and agentic tasks—purpose-built for companies automating engineering workflows and deploying private AI agents.
Qwen3-Coder-30B-A3B-Instruct-AWQ
A 30B MoE coder for private agentic automation: long-context code understanding, tool calling, and repository-scale analysis without leaving your infrastructure.
Devstral-Small-2505-4bit
Lightweight multilingual developer assistant for private deployment in resource-constrained ops environments.
Olmo-3-1125-32B
A 32B fully open-weight base model for companies building private, custom AI systems without proprietary API dependencies.
MiniMax-M1-40k
A 456B parameter open-weight model for enterprises building private, custom AI agents and automating operational workflows without cloud dependency or data residency concerns.
SmolLM2-1.7B-Instruct-GGUF
A 1.7B instruction-tuned model in GGUF format, purpose-built for private deployment on edge hardware and resource-constrained environments where data must stay in-house.
GLM-4.7-Flash-Uncensored-Heretic-NEO-CODE-Imatrix-MAX-GGUF
A private, uncensored 30B MoE model for companies needing unrestricted content generation, creative automation, and reasoning tasks in self-hosted environments.
Qwen2.5-32B-Instruct-bnb-4bit
A 32B quantized instruction-tuned model for private deployment—coding, math, long-context ops tasks, and multilingual automation without vendor lock-in.
Qwen3.6-12B-IQ-Ultra-Heretic-Uncensored-Thinking-V2-Hightop-GGUF
A 12B quantized text model for private deployment in ops workflows—customer data stays on-premises while you automate support, content, and internal knowledge tasks.
Olmo-3-7B-Think
A 7B reasoning model designed for math, coding, and logical inference tasks—deployable privately to automate technical ops workflows without sending reasoning work to external APIs.
Qwen3-0.6B
Mobile-first, quantized LLM for private on-device inference and edge-deployed operational AI—runs full conversations on phones and low-power hardware without calling external APIs.
gpt-oss-120b
120B open-weight MoE for private reasoning agents and custom ops automation—runs on a single 80GB GPU with full chain-of-thought visibility.
Qwen3.6-35B-A3B-Claude-4.6-Opus-Reasoning-Distilled-GGUF
A 35B reasoning-distilled model for self-hosted, chain-of-thought automation—ops teams can run it privately to handle complex analysis, customer support reasoning, and internal decision workflows without external API dependencies.
Qwen2.5-1.5B-Instruct
Compact 1.5B instruction-tuned model for private, resource-constrained ops automation—coding, math, JSON workflows, and long-context tasks on standard infra.
Apertus-70B-Instruct-2509
70B multilingual model designed for private deployment with full compliance transparency—built for ops teams needing data-local LLM infrastructure without proprietary lock-in.
Qwen3-14B-GGUF
14B dense LLM with native switchable reasoning modes—deployable self-hosted for ops automation, agent workflows, and reasoning-heavy tasks without external API dependency.
gpt-oss-20b-speculator.eagle3
Speculative decoding accelerator for gpt-oss-20b: deploy a 20B verifier model faster on your own infrastructure by predicting token sequences in parallel.
Huihui-gpt-oss-20b-BF16-abliterated
20B uncensored base model for private deployments where reduced safety filtering and operational control outweigh standard guardrails.
Qwen3-30B-A3B-quantized.w4a16
A 30B quantized reasoning model (W4A16 INT4) built for cost-efficient private deployment and function-calling ops workflows without sacrificing accuracy.
Llama-3.2-8X3B-MOE-Dark-Champion-Instruct-uncensored-abliterated-18.4B-GGUF
A 18.4B mixture-of-experts creative writing and roleplay engine designed for self-hosted deployment where uncensored, high-volume prose generation and creative task automation matter more than safety guardrails.
LLaDA-1.5
8B diffusion-based LLM optimized for math, code, and alignment tasks—deployable private for ops teams building custom automation without external API dependencies.
Qwen3-14B-Instruct
A 14B instruction-tuned model optimized for finetuning and private deployment, built to automate operational workflows without vendor lock-in.
Qwen3.5-9B-GLM5.1-Distill-v1-GGUF
A 9B distilled reasoning model for private deployment—structured chain-of-thought on a footprint that fits ops infrastructure, trading raw capability for controllable, auditable inference.
TinyLlama_v1.1
A 1.1B parameter Llama-compatible model for cost-conscious private deployments, agent backends, and lightweight operational automation where companies need full data control.
Qwen3.6-35B-A3B-PRISM-NVFP4
Quantized MoE model for private, cost-efficient operational AI on NVIDIA Blackwell hardware, with softened refusal and bias mitigation built in.
North-Mini-Code-1.0-GGUF
Code-focused conversational model for private deployment—run inference on your own infrastructure, integrate into ops workflows, and build custom AI agents without external API dependencies.
Huihui-Qwen3.6-27B-abliterated-NVFP4-MTP
A 27B hybrid-attention reasoning model with built-in speculative decoding, designed to run on 4× consumer/pro GPUs as a private inference backbone for agentic workflows and long-context operational tasks.
open_llama_7b
A permissively licensed 7B reproduction of LLaMA for companies building private, controllable text-generation systems without dependency on Meta's original model.
OLMoE-1B-7B-0924-Instruct
Lightweight MoE backbone for building custom ops automation and private conversational AI without massive infrastructure bills.
Qwen3-0.6B-8bit
Lightweight 0.6B base model for private, on-device text generation and operational automation on edge/consumer hardware.
Ornith-1.0-35B-NVFP4-MTP-GGUF
A 35B Mixture-of-Experts reasoning model optimized for private Blackwell GPU deployment, combining MXFP4/NVFP4 quantization with speculative decoding—built for companies running closed-loop agentic workflows without model exfiltration.
GLM-4.5-Air-FP8
Compact, multi-lingual agentic model (106B total / 12B active) designed for private deployment of reasoning, tool-calling, and code workflows without API dependency.
Qwen2-1.5B-Instruct-GPTQ-Int4
Lightweight instruction-tuned LLM for private deployment in cost-constrained ops environments—automate support, document triage, and internal knowledge workflows without leaving your infrastructure.
PARD-Llama-3.2-1B
A lightweight 1B speculative decoding draft model for accelerating inference in private LLM deployments without target-model retraining.
Seed-OSS-36B-Instruct
A 36B reasoning-optimized foundation model designed for private deployment, custom AI agents, and long-context operational automation in regulated/data-sensitive environments.
Ling-mini-2.0
Sparse MoE model (16B total, 1.4B active) built for cost-efficient private deployment and operational automation without sacrificing reasoning quality.
Qwen3.6-27B-AEON-Ultimate-Uncensored-Multimodal-NVFP4-MTP-XS
27B multimodal reasoning model optimized for high-throughput private deployment on modern GPU clusters; purpose-built for ops automation, agentic workflows, and custom AI applications where inference speed and data residency matter.
grok-1
A large open-weight decoder-LLM in PyTorch, designed for companies running inference on private infrastructure with access to distributed tensor-parallel acceleration.
gemma-4-12B-coder-fable5-composer2.5-v1
A 12B Python coding specialist fine-tuned on verified algorithmic tasks—designed for companies building private code-generation agents and custom AI ops automation on controlled infrastructure.
North-Mini-Code-1.0
Sparse MoE code model (30B/3B active) for private agentic automation—build tool-calling workflows, code generation agents, and terminal tasks that stay in your environment.
plamo-2-1b
A 1B efficient hybrid SSM-attention model for private, bilingual (EN/JA) operational AI—light enough to self-host, not yet instruction-tuned, requiring custom fine-tuning for internal workflows.
Qwen3-1.7B-unsloth-bnb-4bit
1.7B reasoning-capable edge model for private ops automation and lightweight custom AI—thinking/non-thinking toggle for cost-efficient internal tooling.
Phi-4-mini-reasoning
Compact math-reasoning engine for private deployment in ops workflows, knowledge systems, and edge-constrained environments where step-by-step problem-solving must stay in-house.
alias-gpt2-small-x21
A lightweight GPT-2 derivative for companies running private text generation in resource-constrained ops environments where model control and data residency matter more than frontier capability.
snowflake-arctic-instruct
A 480B dense-MoE hybrid for enterprise ops automation and custom AI—built for private deployment with manageable active parameters (17B) and Apache-2.0 freedom.
fg-clip-base
Fine-grained vision-language alignment model for private image classification, retrieval, and multimodal search without vendor lock-in.
Qwen3-8B.w8a8
Quantized 8B reasoning model for cost-efficient private deployment and ops automation without sacrificing inference quality.
Qwen3-4B-Instruct-2507-4bit
A lightweight 4B instruction-tuned model for private deployment on edge/ARM infrastructure; built for ops automation and custom agents in resource-constrained environments.
Qwen2.5-Coder-32B-Instruct-GGUF
A quantized code-generation LLM for self-hosted deployment in engineering ops—automate code review, documentation, and internal tooling without API calls.
Qwen2.5-Coder-14B-Instruct-GGUF
Quantized code-generation model designed for private deployment in ops environments—write, debug, and automate code tasks without vendor lock-in.
Ring-2.5-1T
Trillion-parameter thinking model optimized for deep reasoning and long-horizon agentic workflows in private, self-hosted environments.
Qwen2.5-Coder-14B-Instruct-bnb-4bit
Quantized code-generation LLM for private, self-hosted automation of developer workflows and internal code-assistance tasks.
Qwen3-235B-A22B-Instruct-2507-AWQ
A 235B MoE foundation model optimized for private deployment at scale, delivering competitive performance on reasoning, coding, and multi-language tasks while keeping inference under customer control.
DeepSeek-V3
671B MoE model (37B active) for enterprises building private reasoning agents and automating complex operational workflows without reliance on closed APIs.
Mistral-7B-Instruct-v0.1-GGUF
Quantized 7B instruction-tuned model for CPU/GPU inference in private, air-gapped, or cost-conscious ops environments where you control the weights and data.
Qwen3-4B-DFlash-b16
Speculative decoding drafter for fast, lossless inference acceleration in private Qwen deployments—cut latency by 6x without sacrificing output quality.
Qwen2.5-0.5B-Instruct-GGUF
Ultra-lightweight instruct model (0.5B params) in GGUF format for CPU/edge deployment of chatbots, support automation, and internal agents with full data privacy.
Qwen3-30B-A3B-NVFP4
FP4-quantized 30B MoE base model for cost-efficient private inference and custom ops automation with 75% memory reduction.
Seed-OSS-36B-Instruct-MLX-8bit
36B parameter instruction-tuned model optimized for Apple Silicon private deployment—suitable for companies automating internal workflows without cloud dependency.
Olmo-Hybrid-7B
A 7B hybrid RNN-transformer model optimized for long-context inference efficiency and data-efficient training—designed for companies needing cost-effective private deployment with 2x the throughput of comparable dense transformers.
Qwen3.6-14B-A3B-VibeForged-v2-GGUF
A 14B quantized MoE reasoning model optimized for local deployment, custom fine-tuning, and private inference—built to run on modest hardware while maintaining multimodal capabilities.
GLM-5-NVFP4
A quantized 435B-parameter mixture-of-experts model optimized for private inference on NVIDIA hardware—designed for teams building custom AI agents, internal knowledge systems, and workflow automation that must stay in their own infrastructure.
Qwen3-0.6B-GGUF
Lightweight reasoning model (0.6B) for private ops automation and edge deployment; toggles between thinking mode (math/code) and fast inference mode.
pythia-410m-deduped
Research-grade 410M base model for building interpretable, self-hosted text-generation systems and automating content workflows without proprietary dependencies.
Qwen3-8B-quantized.w4a16
INT4-quantized 8B reasoning model optimized for private deployment—75% smaller footprint, full Apache 2.0 freedom, production-ready for ops workflows.
granite-4.0-micro
3B instruction-tuned model purpose-built for enterprise ops automation, tool-calling agents, and private deployment in mid-market companies.
OpenThinker2-7B
A 7B reasoning model optimized for math, code, and complex problem-solving—deployable privately to automate analytical workflows and custom reasoning applications without external API dependency.
Qwen3.5-27B-DFlash
A lightweight speculative-decoding draft model designed to accelerate Qwen3.5-27B inference in production serving—not standalone, but a throughput multiplier for ops-heavy workloads when paired with a target model.
prometheus-7b-v2.0
A specialized evaluation LLM for scoring and ranking AI outputs—designed for ops teams automating quality assessment workflows in private environments.
Moonlight-16B-A3B-Instruct
A 16B-parameter Mixture-of-Experts model that activates only 3B parameters per inference, designed for cost-efficient private deployment in ops automation and custom AI applications where inference speed and resource footprint matter.
A.X-K1
A sparse 519B-parameter MoE model with 33B active parameters, designed for hybrid reasoning/fast-inference workflows in private, self-hosted deployments where cost-per-inference and reasoning depth are both controllable.
Qwen2.5-1.5B-Instruct-GGUF
A 1.5B instruction-tuned model in GGUF format—lightweight enough to run on consumer hardware, production-ready for private deployment across customer environments.