← Coding agents & IDE assistants
MTPLX
The fastest way to run Qwen 3.8 Flash Next, Qwen 3.8 27B and Ternary Bonsai 2 27B on a Mac: 125 tok/s in OpenCode on an M5 Max, and a 27B model on 16 GB Macs. Native MTP speculative decoding on Apple Silicon, exact at any temperature. OpenAI and Anthropic compatible local server.
Ranked by AgentPick Score, our data score from public GitHub activity. Description is the project's own.
Topics: anthropic-compatible, apple-silicon, claude-code, flash-next, inference-engine, llm-inference, local-ai, local-llm, macos, metal, mlx, mtp
Starter prompt
Using MTPLX, in a throwaway branch: add a unit test for [function] covering [edge case]. Don't change production code. Show me the diff and the test output.
More in Coding agents & IDE assistants
Open Interpreter →
A coding agent for open models like Kimi K3 and GLM 5.3
oh-my-pi →
⌥ Coding agent with the IDE wired in. Built by Stencil Labs.
Hermes Agent →
The agent that grows with you
Codewhale →
Open-source Rust agent engine and terminal client for Codewhale, with provider choice, tools, approvals and re…
Qwen Code →
An open-source AI coding agent that lives in your terminal.
DeepSeek-Reasonix →
A reliable coding agent for complex software engineering tasks.
AgentPick Weekly
The week's new agent tools with a verdict on each, every week. Free. We'll email you a link to confirm, and you can unsubscribe any time. See our Privacy Policy.