# MiniMax-AI/MiniMax-M2

MiniMax-M2, a model built for Max coding & agentic workflows.

Repository: https://github.com/MiniMax-AI/MiniMax-M2
Canonical: https://ross.abutalabs.com/products/minimax-m2
Homepage: https://www.minimax.io/
License: NOASSERTION
License Family: other
Topics: large-language-models, llm
Last push: 2025-11-13T08:12:36+00:00

## Health v2 (maintenance only)
Score: 40/100 (v2, computed 2026-09-03T02:20:16.233290+00:00)
- activity 52, release rhythm 35, longevity 22
- inputs: {"age_days": 312, "days_push": 293, "days_rel": null, "gap_med": null, "n_releases_24m": 0}
- flags: no_releases, no_license
- formula: round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10)

## Adoption (not part of the score)
Stars 2605, forks 215 (observed 2026-08-28T04:07:04.019407+00:00)

## What it is
MiniMax-M2 is an open-source Mixture-of-Experts large language model with 230 billion total parameters and 10 billion active parameters, released by MiniMax for coding and agentic workflows. The repository provides model weights, license information, and documentation for running the model locally or via the MiniMax API.

## Use cases
- run an open-source LLM for coding agents
- self-host a cost-effective MoE model for tool use
- power multi-file code editing and coding-run-fix loops
- build agentic workflows with strong tool-calling performance
- serve an LLM locally for code generation
- evaluate open-source models on coding and agentic benchmarks

## When to choose
- you need an open-weight model optimized for coding and agent tool use
- you want near-frontier agentic performance with only 10B active parameters for cheaper inference
- you want to self-host or fine-tune a top-ranked open-source coding model

## When to avoid
- you need multimodal input like images or video
- you need a small model that fits on consumer hardware
- you only need a hosted API without self-hosting

## Facets
- artifact type: dataset
- maturity: active
- function: llm-inference, machine-learning, agent-framework
- domain: large-language-models, artificial-intelligence, machine-learning
- platform: python, self-hosted, cloud
- tags: open-weights, mixture-of-experts, model-weights, coding-llm, agentic-workflows, minimax, huggingface, ai-agents, gpu

## Member repositories
- MiniMax-AI/MiniMax-M2 (main) score 40

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:07:04.019407+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 2026-08-30T02:21:01.537601+00:00, confidence not recorded.
  - readme: https://github.com/MiniMax-AI/MiniMax-M2 (fetched 2026-08-28T04:07:04.019407+00:00, sha 9fbde13445d5)
  - homepage: https://www.minimax.io/ (fetched 2026-08-29T10:04:21.245561+00:00, sha 4750f71d0331)
  - site_page: https://www.minimax.io/about (fetched 2026-08-29T10:04:21.254807+00:00, sha dea11ef0f24f)
  - site_page: https://platform.minimax.io/docs/api-reference/video-generation-v2-create (fetched 2026-08-29T10:04:21.256768+00:00, sha 75d5cfe870c0)
  - site_page: https://platform.minimax.io/docs/guides/text-generation (fetched 2026-08-29T10:04:21.258719+00:00, sha 260a4f53a7ef)
  - site_page: https://platform.minimax.io/docs/guides/video-generation (fetched 2026-08-29T10:04:21.260489+00:00, sha 7c0b7ec5a142)
  - site_page: https://platform.minimax.io/docs/guides/speech-voice-clone (fetched 2026-08-29T10:04:21.262366+00:00, sha 656927527880)
- Data as of 2026-08-30T08:39:29.467469+00:00.