AI & ML interests
Open-source OpenAI- and Anthropic-compatible LLM runtime for Apple Silicon. Up to 4× faster than mlx-lm, same weights. Reliable tool calling for coding agents.
Recent Activity
View all activity
Organization Card
Rapid-MLX
Local OpenAI- and Anthropic-compatible inference for Apple Silicon.
Run it on your Mac
curl -fsSL https://rapidmlx.com/install.sh | bash
rapid-mlx serve ling-3.0-tiny-4bit
curl -s http://localhost:8000/v1/chat/completions -H 'Content-Type: application/json' -d '{"model":"default","messages":[{"role":"user","content":"Say hello"}]}'The example needs at least 8 GB unified memory; allow more for long contexts.
Use it from OpenCode / Claude Code / Codex: rapid-mlx agents <name> --setup · Agent setup
models 17
rapid-mlx/NVIDIA-Nemotron-3.5-Bolt-E4B-MLX-8bit
Text Generation • 7B • Updated • 259
rapid-mlx/NVIDIA-Nemotron-3.5-Bolt-E4B-MLX-4bit
Text Generation • 7B • Updated • 241
rapid-mlx/DeepSeek-V4.1-Flash-1017-MLX
Text Generation • 270B • Updated • 2.63k • 4
rapid-mlx/NVIDIA-Nemotron-3.5-Super-VL-120B-A12B-MLX-3bit
Text Generation • 121B • Updated • 1.03k
rapid-mlx/NVIDIA-Nemotron-3.5-Super-VL-120B-A12B-MLX-4bit
Text Generation • 121B • Updated • 1.04k • 1
rapid-mlx/NVIDIA-Nemotron-3.5-Bolt-74B-A9B-MLX-4bit
Text Generation • 74B • Updated • 1.03k • 1
rapid-mlx/NVIDIA-Nemotron-3.5-Bolt-74B-A9B-MLX-3bit
Text Generation • 74B • Updated • 1.3k • 2
rapid-mlx/Qwen3.8-Flash-Next-4bit
Text Generation • 76B • Updated • 1.26k • 5
rapid-mlx/Qwen3.8-27B-mixed-3.5bpw-MLX
Image-Text-to-Text • 27B • Updated • 687 • 10
rapid-mlx/Qwen3.8-27B-4bit-MTP-fp16-MLX
Text Generation • 27B • Updated • 277
datasets 0
None public yet