mtplx 开源项目推荐（14 个）

youssofal /MTPLX

在 Qwen 3.6 27B（温度 0.6）上，解码 TPS 提升 2.24 倍 | 在 Apple Silicon 上无需外部 Drafter 的原生 MTP 推测解码。

KrossKinetic /pi-mtplx

Run local MLX models with Pi — zero config, automatic model discovery, and live token-speed monitoring.

Upinel /UpinelAIOS-MLX

One-click uncensored AI agent server for Apple Silicon Macs. MLX + MTPLX speculative decoding, OpenAI-compatible on your LAN. Up to 79.4 t/s decode, MTP speculative decoding on.

Upinel /AIOS

One-click uncensored AI agent server for Apple Silicon Macs. OpenAI-compatible on your LAN. Gemma 4 and Qwen. Up to 239 t/s decode. (GGUF,MLX,Splash)

spinloop-ai /spinloop

CLI to deploy and control a fleet of LLMs on local machine / LAN / Tailscale or AWS. Spin up your agent loop.

alexvnesta /claude-science-local-model-lab

Run an isolated Claude Science app copy through local or OpenAI-compatible model backends.

devty /mtplx-dashboard

Beautiful, zero-dependency realtime dashboard + live activity log for a local MTPLX inference server — reads the /metrics endpoint, no build step.

jtc268 /balto-speedrunner-mac

Double Qwen 3.8 27B inference speed on Apple Silicon with a one-click local coding agent

charlesnutter /opencode-headsup

A heads-up display with per-turn/session telemetry for both local engine and cloud models. Data straight from the engine. Look for it in the sidebar.

AlexanderMarzollDev /mtplx-m5max-64gb-benchmarks

Local LLM benchmarks on a 64GB Mac M5 Max: mtplx with Qwen3.8-27B and Qwen3.6-35B-A3B (MTP, long context, engines, settings)

saman-mb /better-mtplx-ui

A lightweight, parallel-first desktop GUI for MTPLX — browse packs, launch multiple models on their own ports, watch them side by side.

TobyNoSkillSon /MTPLX-Switchboard

Native macOS menu-bar controller for local MTPLX model daemons.

Orbital-ML-Lab /qwen-fp8-mtp-sidecar-for-mtplx

Recover the native MTP predictor missing from the 8-bit MLX Qwen3.8-27B-Uncensored package, build a BF16 sidecar, and reproduce a 15.59 → 48.75 tok/s controlled M4 Max result with MTPLX.

BartRuSec /llm-runtime-dock

Local OpenAI/Anthropic-compatible LLM gateway that starts, stops and switches local runtimes (MTPLX, LM Studio, oMLX, Ollama) behind one stable endpoint for your coding agents.