MiniMax-M2: pricing, context window and benchmarks
MiniMax-M2 is a multimodal model from MiniMax. It offers a 200,000-token context window, costs $5.19 per million input tokens and $22.73 per million output tokens, and averages 80.6 across the index's benchmark suite.
Pricing
MiniMax-M2 costs $5.19 per million input tokens and $22.73 per million output tokens. A blended 3:1 input-to-output mix works out to about $9.57 per million tokens. Cached input is listed at $0.50 per million tokens.
| Item | Price | Notes |
|---|---|---|
| Input | $5.19 | per 1M tokens |
| Output | $22.73 | per 1M tokens |
| Cached input | $0.50 | per 1M tokens |
| Blended 3:1 | $9.57 | estimate per 1M tokens |
| 1M in + 1M out | $27.92 | estimate |
Benchmarks
Scores below are modelled estimates on a 0–100 scale. Use them to compare relative position, not as official vendor results.
| Benchmark | Score | Rank | Percentile | Relative |
|---|---|---|---|---|
| MMLU-Pro | 81.8 | 127 | 79th | |
| GPQA Diamond | 75.4 | 137 | 77th | |
| SWE-bench Verified | 69.9 | 112 | 81th | |
| HumanEval | 88.0 | 73 | 88th | |
| MATH-500 | 82.3 | 157 | 74th | |
| AIME 2025 | 78.1 | 115 | 81th | |
| LiveCodeBench | 73.4 | 121 | 80th | |
| IFEval | 86.1 | 153 | 75th | |
| TruthfulQA | 83.5 | 93 | 85th | |
| HellaSwag | 90.4 | 76 | 87th | |
| MMMU | 78.0 | 98 | 84th | |
| Arena Elo | 1423.8 | 130 | 78th |
What MiniMax-M2 is good for
MiniMax-M2 is tuned toward coding work. It is a reasonable candidate for code generation, review, debugging and repository-scale tasks, especially when paired with a retrieval layer. At $5.19 per million input tokens, it sits in the mid-range of the index, and the 200,000-token context window is enough for large files and multi-file diffs.
Strengths
- Large 200,000-token context window for long documents and agent traces.
- Open weights make self-hosting, fine-tuning and data-residency review possible.
- Accepts more than text, including image input where the provider supports it.
- Coding benchmarks are a relative strength within its tier.
Trade-offs
- No standout weakness in the indexed specifications, but real-world behaviour still needs testing.
How it compares
| Model | Provider | Context | Input / 1M | Avg score |
|---|---|---|---|---|
| Kimi K1.5 | Moonshot AI | 128000 | $5.19 | 80.7 |
| ERNIE 4.5 Max | Baidu | 128000 | $14.41 | 80.6 |
| MiniMax-M2 | MiniMax | 200000 | $5.19 | 80.6 |
| ERNIE 4.5 Pro | Baidu | 1000000 | $13.89 | 80.5 |
| Kimi K1.5 Instruct | Moonshot AI | 200000 | $5.15 | 80.5 |
| Sonar Pro Reasoning | Perplexity | 128000 | $13.94 | 80.5 |
FAQ
How much does MiniMax-M2 cost?
MiniMax-M2 is listed at $5.19 per million input tokens and $22.73 per million output tokens. Cached input is $0.50 per million tokens. A blended 3:1 input-to-output mix is about $9.57 per million tokens.
What is the context window of MiniMax-M2?
MiniMax-M2 has a context window of 200,000 tokens and can emit up to 65,536 output tokens per request in this index.
Is MiniMax-M2 open weights?
Yes. MiniMax-M2 is listed as an open-weight model under a open weights licence.
How does MiniMax-M2 score on benchmarks?
MiniMax-M2 averages 80.6 across the scored benchmarks in this index, ranking #105 of 600 models. Its strongest relative result is Arena Elo at 1423.8.
What is MiniMax-M2 best for?
MiniMax-M2 is tuned toward coding work. It is a reasonable candidate for code generation, review, debugging and repository-scale tasks, especially when paired with a retrieval layer. At $5.19 per million input tokens, it sits in the mid-range of the index, and the 200,000-token context window is enough for large files and multi-file diffs.