# MiniMax-AI/MiniMax-M1

MiniMax-M1, the world's first open-weight, large-scale hybrid-attention reasoning model.

Repository: https://github.com/MiniMax-AI/MiniMax-M1
Canonical: https://ross.abutalabs.com/products/minimax-m1
Homepage: https://www.minimax.io/
Language: Python
License: Apache-2.0
License Family: permissive
Topics: large-language-models, llm, reasoning-models, minimax-m1
Last push: 2025-07-07T11:57:22+00:00

## Health v2 (maintenance only)
Score: 32/100 (v2, computed 2026-09-02T17:46:02.011165+00:00)
- activity 30, release rhythm 35, longevity 31
- inputs: {"age_days": 445, "days_push": 422, "days_rel": null, "gap_med": null, "n_releases_24m": 0}
- flags: no_releases
- formula: round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10)

## Adoption (not part of the score)
Stars 3180, forks 283 (observed 2026-08-28T04:07:47.777186+00:00)

## What it is
MiniMax-M1 is an open-weight, large-scale hybrid-attention reasoning language model released by MiniMax under Apache-2.0. The repository provides model weights, inference examples, and deployment guidance for running the model locally.

## Use cases
- run an open-weight reasoning llm locally
- deploy a long-context language model on my own gpu server
- fine-tune or evaluate a hybrid-attention reasoning model
- self-host an apache-licensed alternative to closed reasoning models
- serve a large language model for agentic workflows

## When to choose
- you need open weights and permissive licensing for a reasoning model
- you want to self-host inference instead of calling a paid api
- you need long-context reasoning capabilities you can inspect and modify

## When to avoid
- you only want a hosted api without managing gpu infrastructure
- you lack the hardware to run a large-scale model
- you need multimodal video or speech generation rather than text reasoning

## Facets
- artifact type: library
- maturity: active
- function: llm-inference, machine-learning, deep-learning
- domain: large-language-models, artificial-intelligence, deep-learning
- platform: python
- tags: open-weights, reasoning-model, hybrid-attention, long-context, transformers, gpu, linux

## Member repositories
- MiniMax-AI/MiniMax-M1 (main) score 32

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:07:47.777186+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 2026-08-30T07:25:05.106717+00:00, confidence not recorded.
  - readme: https://github.com/MiniMax-AI/MiniMax-M1 (fetched 2026-08-28T04:07:47.777186+00:00, sha 37951652c6cf)
  - homepage: https://www.minimax.io/ (fetched 2026-08-29T09:39:17.801159+00:00, sha 4750f71d0331)
  - site_page: https://www.minimax.io/about (fetched 2026-08-29T09:39:17.803754+00:00, sha dea11ef0f24f)
  - site_page: https://platform.minimax.io/docs/api-reference/video-generation-v2-create (fetched 2026-08-29T09:39:17.805884+00:00, sha 75d5cfe870c0)
  - site_page: https://platform.minimax.io/docs/guides/text-generation (fetched 2026-08-29T09:39:17.808707+00:00, sha 260a4f53a7ef)
  - site_page: https://platform.minimax.io/docs/guides/video-generation (fetched 2026-08-29T09:39:17.811402+00:00, sha 7c0b7ec5a142)
  - site_page: https://platform.minimax.io/docs/guides/speech-voice-clone (fetched 2026-08-29T09:39:17.813832+00:00, sha 656927527880)
- Data as of 2026-08-30T08:39:29.467469+00:00.