ai/llama3.2

Docker

Solid LLaMA 3 update, reliable for coding, chat, and Q&A tasks

ai/deepseek-r1-distill-llama

Docker

Distilled LLaMA by DeepSeek, fast and optimized for real-world tasks

ai/llama3.3

Docker

Newest LLama 3 release with improved reasoning and generation quality

ai/smollm3

Docker

SmolLM3 is a 3.1B model for efficient on-device use, with strong performance in chat

ai/kimi-k2

Docker

Kimi K2 Thinking: open-source agent with deep reasoning, stable tool use, fast INT4, 256k context.

ai/stable-diffusion

Docker

Image generation model, uses a base latent diffusion model plus a refiner.

ai/qwen2.5

Docker

Versatile Qwen update with better language skills and wider support

ai/llmman

Docker

llmman with llama.cpp: server, server-cuda, server-cuda13, server-rocm, server-vulkan

ai/ministral3-vllm

Docker

Ministral 3: compact vision-enabled model with near-24B performance, optimized for local edge use

ai/granite-docling

Docker

Granite Docling is a multimodal model for efficient document conversion.

ai/deepseek-v4-flash-safetensors

Docker

Efficient 284B MoE language model with 1M token context and multi-mode reasoning capabilities

ai/gemma3-qat

Docker

Google’s latest Gemma, in its QAT (quantization aware trained) variant

ai/phi4

Docker

Microsoft’s compact model, surprisingly capable at reasoning and code

ai/kimi-k2.6

Docker

1T MoE multimodal agentic model with long-horizon coding, swarm orchestration, and native vision

ai/ministral3

Docker

Ministral 3: compact vision-enabled model with near-24B performance, optimized for local edge use

ai/qwen3-coder-next

Docker

Efficient 80B MoE coding model with 3B activated params, 256K context, and agentic capabilities