Model List

The available models change over time as new models are added and older ones are deprecated. For tasks that only require a general LLM or embedding model, use the provided tags instead of hard-coding specific model names. Tags are fully supported as model identifiers in the API and offer a more stable interface, even as the underlying models are updated. The "best-*" tag is a special case that always points to the best model for a given task.

Models are provided on a best-effort basis, which means they will usually be available, but occasional interruptions may occur.

Large Language Models

NameModelDocumentationLicenseTags
best-chatopenai/gpt-oss-120bhttps://huggingface.co/openai/gpt-oss-120bApache 2.0
best-chat
chat
best-codeqwen/qwen3-coder-30bhttps://huggingface.co/Qwen/Qwen3-Coder-30B-A3B-InstructApache 2.0
best-code
chat
best-visiongemma-3-27b-ithttps://huggingface.co/mlx-community/gemma-3-27b-it-8bitGemma
best-vision
vision
granite-docling-258m-mlxgranite-docling-258m-mlxhttps://huggingface.co/ibm-granite/granite-docling-258M-mlxApache 2.0
chat
vision
gpt-oss-120bopenai/gpt-oss-120bhttps://huggingface.co/openai/gpt-oss-120bApache 2.0
best-chat
chat
minimax/minimax-m2minimax/minimax-m2https://huggingface.co/mlx-community/MiniMax-M2-4bitmit
chat
qwen3-vl-30b:2qwen/qwen3-vl-30b:2https://huggingface.co/lmstudio-community/Qwen3-VL-30B-A3B-Thinking-MLX-8bitApache 2.0
chat
vision
gpt-oss-20bopenai/gpt-oss-20bhttps://huggingface.co/openai/gpt-oss-20bApache 2.0
chat
vibethinker-1.5b-mlxvibethinker-1.5b-mlxhttps://huggingface.co/WeiboAI/VibeThinker-1.5Bmit
chat
qwen3-coder-30bqwen/qwen3-coder-30bhttps://huggingface.co/Qwen/Qwen3-Coder-30B-A3B-InstructApache 2.0
best-code
chat
gemma-3-27b-itgemma-3-27b-ithttps://huggingface.co/mlx-community/gemma-3-27b-it-8bitGemma
best-vision
vision
qwen3-vl-30bqwen/qwen3-vl-30bhttps://huggingface.co/lmstudio-community/Qwen3-VL-30B-A3B-Thinking-MLX-8bitApache 2.0
chat
vision

Embedding Models

NameModelDocumentationLicenseTags
best-embeddingintfloat/multilingual-e5-large-instructhttps://huggingface.co/intfloat/multilingual-e5-large-instructmit
best-embedding
embedding
nomic-embed-text-v1.5nomic-ai/nomic-embed-text-v1.5https://huggingface.co/nomic-ai/nomic-embed-text-v1.5Apache 2.0
embedding
multilingual-e5-large-instructintfloat/multilingual-e5-large-instructhttps://huggingface.co/intfloat/multilingual-e5-large-instructmit
best-embedding
embedding
bge-reranker-v2-m3 text-embedding-bge-reranker-v2-m3https://huggingface.co/BAAI/bge-reranker-v2-m3#model-listApache 2.0
embedding