MiniMax-AI/MiniMax-M1

Health v2 · maintenance only

Flags: no_releases

Adoption not part of the score

3180 stars · 283 forks observed · 2026-08-28

What it is AI-extracted, prompt v1, taxonomy v1, 2026-08-30, confidence not recorded

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

library · maturity active

llm-inference machine-learning deep-learning large-language-models artificial-intelligence deep-learning python open-weights reasoning-model hybrid-attention long-context transformers gpu linux

Member repositories

For agents

markdown · JSON · MCP: product_card(name="MiniMax-AI/MiniMax-M1")