MiniMax M2: Benchmarks, Pricing & Context Window
MiniMax M2 is a language model from MiniMax, released in October 2025, with a 1M-token context window, and pricing from $0.300/M input and $1.20/M output.
MiniMax M2 is an open-source large language model by MiniMax, built for agents and coding tasks. It delivers state-of-the-art tool use, reasoning, and search performance while maintaining exceptional cost-efficiency and speed, priced at
MiniMax M2 benchmarks
Capability tiers
Standing within each category, adjusted for leaderboard depth.
Real tasks performance
High-confidence performance for MiniMax M2 across real-world prompt categories. Only 95% intervals at most 4 points wide are shown.
Performance by conversation depth
How MiniMax M2 holds up as conversations get longer.
Quality Tracker
MiniMax M2 Performance Across Datasets
Scores sourced from the model's scorecard, paper, or official blog posts
llm-stats.com - Mon Sep 14 2026MiniMax M2 pricing
Providers
MiniMax M2 starts at $0.300 per million input tokens and $1.20 per million output tokens via MiniMax. See all 2 providers below with their per-token pricing, latency, throughput, and modality support.
| Provider | Input $/M | Cached input $/M | Output $/M | Context in / out | TTFT p95 s | Output p5 c/s | Modalities in / out |
|---|---|---|---|---|---|---|---|
 MiniMax | $0.300 | — | $1.20 | 1.0M/1.0M | 4.00 | — | / |
 Novita | $0.300 | $0.0300 | $1.20 | 204.8K/131.1K | — | — | / |
Cached input is the discounted price for prompt tokens served from a provider cache. TTFT is time to first token. Output is characters per second; p5 is the sustained floor exceeded by 95% of observed requests.
MiniMax M2 model size
MiniMax M2 has 230 billion parameters. See how it compares to other models in the same parameter range.
MiniMax M2 context window
Input and output token limits for MiniMax M2, plus how it ranks on long-context understanding.
Try now
huggleMake it with
MiniMax M2.
MiniMax M2 latency
MiniMax M2 time to first token, sustained output throughput, and failed-request rate from live model usage over the trailing 7 days.
MiniMax M2 examples
Recent arena outputs from MiniMax M2, picked from the highest-ranked matchups.
MiniMax M2 license
MiniMax M2 is released under the MIT license, which permits commercial use, has 230.0B parameters.
- License
- MIT
- Commercial use allowed
- Parameters
- 230.0B
MIT License - allows commercial use
MiniMax M2 resources
Official sources for MiniMax M2: provider documentation, official launch post, source repository, model weights.
MiniMax M2 vs other models
The most-compared alternatives to MiniMax M2 are Phi 4 Reasoning Plus, Qwen3 VL 235B A22B Instruct, Qwen3 VL 32B Thinking. Open any pair side-by-side for benchmarks, pricing, context, and latency.
Models like MiniMax M2
Models ranked just above and below MiniMax M2 by LLM Stats score.
FAQ
Common questions about MiniMax M2.