# BiFangKNT/mtga

基于本地代理的方式，绕过 IDE 的固定模型服务商限制

Repository: https://github.com/BiFangKNT/mtga
Canonical: https://ross.abutalabs.com/products/mtga
Language: Python
License: AGPL-3.0
License Family: copyleft
Last push: 2026-06-18T11:51:22+00:00

## Health v2 (maintenance only)
Score: 76/100 (v2, computed 2026-09-02T17:46:02.011165+00:00)
- activity 88, release rhythm 84, longevity 34
- inputs: {"age_days": 476, "days_push": 76, "days_rel": 107, "gap_med": 6, "n_releases_24m": 24}
- flags: none
- formula: round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10)

## Adoption (not part of the score)
Stars 1179, forks 128 (observed 2026-08-28T04:03:53.251001+00:00)

## What it is
MTGA is a local proxy application that lets IDEs with fixed model providers (such as Trae) use arbitrary LLM backends. It exposes an OpenAI Chat Completions-compatible API downstream and forwards requests to upstream providers like OpenAI, Anthropic, and Gemini via a built-in MLiteLLM execution layer.

## Use cases
- use custom LLM providers in an IDE that locks you to one model vendor
- route IDE AI requests to Anthropic or Gemini through an OpenAI-compatible endpoint
- run a local proxy to bypass fixed model provider restrictions in Trae IDE
- convert between OpenAI chat completions and responses APIs at a proxy layer
- self-host a lightweight LLM API forwarding layer without third-party frameworks

## When to choose
- you use an IDE like Trae that restricts which model providers you can configure
- you want a single local proxy exposing an OpenAI-compatible API to multiple upstream LLM providers
- you prefer a self-contained tool with no extra forwarding framework dependencies

## When to avoid
- you need a general-purpose production API gateway rather than an IDE-focused proxy
- you require Linux support, which is not listed among supported platforms
- you want a managed cloud service instead of a locally-run proxy

## Facets
- artifact type: application
- maturity: active
- function: proxy, llm-inference, api-framework, http-server, developer-tools
- domain: developer-tools, large-language-models, apis, self-hosted
- platform: windows, python, self-hosted
- tags: local-proxy, ide-model-bypass, openai-compatible-api, trae-ide, api-forwarding, litellm, macos

## Member repositories
- BiFangKNT/mtga (main) score 76

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:03:53.251001+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 2026-08-30T06:25:26.009315+00:00, confidence not recorded.
  - readme: https://github.com/BiFangKNT/mtga (fetched 2026-08-28T04:03:53.251001+00:00, sha 0f626019e367)
- Data as of 2026-08-30T08:39:29.467469+00:00.