# raullenchai/Rapid-MLX

The fastest local AI engine for Apple Silicon. 4.2x faster than Ollama, 0.08s cached TTFT, 100% tool calling. 17 tool parsers, prompt cache, reasoning separation, cloud routing. Drop-in OpenAI replacement. Works with Claude Code, Cursor, Aider.

Repository: https://github.com/raullenchai/Rapid-MLX
Canonical: https://ross.abutalabs.com/products/rapid-mlx
Homepage: https://pypi.org/project/rapid-mlx
Language: Python
License: NOASSERTION
License Family: other
Topics: apple-silicon, fastapi, inference, llm, local-llm, macos, mlx, openai-api, python, tool-calling, hacktoberfest, ollama-alternative, m1, m2, m3, qwen, deepseek, claude-code, cursor
Last push: 2026-08-26T22:50:20+00:00

## Health v2 (maintenance only)
Score: 78/100 (v2, computed 2026-09-03T02:39:23.370411+00:00)
- activity 99, release rhythm 87, longevity 13
- inputs: {"age_days": 190, "days_push": 7, "days_rel": 7, "gap_med": 0, "n_releases_24m": 218}
- flags: no_license
- formula: round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10)

## Adoption (not part of the score)
Stars 3551, forks 401 (observed 2026-08-28T04:08:09.556474+00:00)

## What it is
Rapid-MLX is a local LLM inference engine for Apple Silicon Macs built on MLX, exposing a drop-in OpenAI/Anthropic-compatible API. It offers fast throughput, prompt caching, tool calling, and cloud routing as an Ollama alternative.

## Use cases
- run local llm on mac m1 m2 m3
- openai-compatible local inference server
- faster alternative to ollama on apple silicon
- local backend for claude code or cursor
- serve llm with tool calling locally
- self-host llm api on macbook

## When to choose
- you want fast local LLM inference on an M-series Mac
- you need an OpenAI-compatible endpoint for coding agents like Claude Code, Cursor, or Aider
- you want prompt caching and reliable tool calling locally

## When to avoid
- you need inference on Linux, Windows, or NVIDIA GPUs
- you require a fully permissive license for redistribution (license is non-standard)
- you need multi-node or cluster-scale serving

## Facets
- artifact type: service
- maturity: active
- function: llm-inference, http-server, api-framework, prompt-engineering, agent-framework
- domain: large-language-models, artificial-intelligence, developer-tools, self-hosted
- platform: python, self-hosted, cli
- tags: apple-silicon, mlx, ollama-alternative, openai-compatible-api, tool-calling, local-llm, prompt-cache, claude-code, cursor, macos

## Member repositories
- raullenchai/Rapid-MLX (main) score 78

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:08:09.556474+00:00.
- Health v2: computed from the inputs above; adoption is never an input.
- Inferred fields (summary, facets, guidance): AI-extracted, prompt v1, taxonomy v1, on 2026-08-29T18:34:20.391807+00:00, confidence not recorded.
  - readme: https://github.com/raullenchai/Rapid-MLX (fetched 2026-08-28T04:08:09.556474+00:00, sha efd85f414583)
  - homepage: https://pypi.org/project/rapid-mlx (fetched 2026-08-29T09:28:17.330172+00:00, sha 4b4e8fead74a)
  - registry_pypi: https://pypi.org/pypi/rapid-mlx/json (fetched 2026-08-29T09:28:17.332650+00:00, sha b755b69a57d5)
- Data as of 2026-08-30T08:39:29.467469+00:00.