Local LLM model page

MiniMax M2 (230B MoE)

MiniMax MoE flagship with 10B active params and 4M-token long-context. Specialised for agentic coding and tool-use. Competitive with GPT-4 class models at a fraction of the inference cost. MIT licensed.

Find the best model for my hardwareBrowse all 183 LLMs
Parameters
230B (10B active)
Minimum RAM
192 GB
Model size
140 GB
Quantization
Q4_K_M

Can MiniMax M2 (230B MoE) run locally?

MiniMax M2 (230B MoE) is best suited for server-grade or multi-GPU systems. LocalClaw recommends Q4_K_M as the default quantization, with at least 192 GB RAM.

Search term for LM Studio or compatible runtimes: minimax-m2

Hugging Face repository: MiniMaxAI/MiniMax-M2

chatcodereasoningquality

Strengths

  • MiniMax MoE flagship with 10B active params and 4M-token long-context. Specialised for agentic coding and tool-use. Competitive with GPT-4 class models at a fraction of the inference cost. MIT licensed.

Limitations

  • Performance depends heavily on quantization, RAM bandwidth and runtime support.

Best use cases

  • chat
  • code
  • reasoning
  • quality

Benchmarks

Speed: 5/10

Quality: 9/10

Coding: 10/10

Reasoning: 9/10

Technical details

Developer: minimax

License: See model repository

Context window: Unknown tokens

Architecture: See model card

Released: 2025-10