See key specs and per-benchmark scores for each model/mode. Scroll horizontally for all columns. 当前对比 2 个模型的评测数据与核心参数。

MiniMax-M2.7
MiniMaxAI
Each axis is a category average, normalized to a 100-point radar.
Relative edge: 生产力知识 +14.0 / Relative gap: Agent能力评测 -12.8
Relative edge: Agent能力评测 +12.8 / Relative gap: 生产力知识 -14.0
Method: for each model and benchmark, the chart first averages all scores in the current mode scope instead of taking the best score, then averages those benchmark scores within each category. Only benchmarks with at least two selected models scored are included; missing values are not counted as zero.
Best overall
MiniMax-M2.7 · 70.00
Best single
MiniMax M2.5 · τ²-Bench - Telecom 97.80
Modality coverage
MiniMax-M2.7 · 0 modalities
Head to head
9
Benchmarks
5
Wins
4
Losses
+1.87
Average diff
Compare benchmark results across thinking modes and tool usage.
Data sourced primarily from official releases (GitHub, Hugging Face, papers), then benchmark leaderboards, then third-party evaluators. Learn about our data methodology
Complete scores for each model/mode across selected benchmarks.
9 benchmarks with comparable scores. Each model shows its best score; mode label is displayed below.
| Benchmark | MiniMax-M2.7 | MiniMax M2.5 |
|---|---|---|
GPQA Diamond 综合评估 | 87.00Thinking Enabled | 85.20Thinking Enabled |
HLE 综合评估 | 28.00Thinking Enabled | 19.40Thinking Enabled |
SWE-Bench Pro - Public 编程与软件工程 | 56.20Thinking Enabled | Tools | 55.40Thinking Enabled | Tools |
τ²-Bench - Telecom Agent能力评测 | 85.00Thinking Enabled | Tools | 97.80Thinking Enabled | Tools |
IF Bench 指令跟随 | 76.00Thinking Enabled | Tools | 70.00Thinking Enabled | Tools |
GDPval-AA 生产力知识 | 50.00Thinking Enabled | 36.00Thinking Enabled |
AA-LCR 长上下文能力 | 69.00Thinking Enabled | Tools | 69.50Thinking Enabled |
Claw Bench OpenClaw智能体能力综合测评 | 91.70Thinking Enabled | Tools | 92.10Thinking Enabled | Tools |
Pinch Bench OpenClaw智能体能力综合测评 | 87.10Thinking Enabled | Tools | 87.80Thinking Enabled | Tools |
Side-by-side input/output token pricing
Licensing, MoE architecture, and multi-modality support.
| Features & specs | MiniMax-M2.7MiniMaxAI | MiniMax M2.5MiniMaxAI |
|---|---|---|
Core specsRelease | 2026-03-18 | 2026-02-12 |
Context length | 200K | 128K |
Parameters | 2290 | 2290 |
Active parameters | 100 | 100 |
Max output | 204800 | Not provided |
MoE | Yes | Yes |
LicenseCode Open Source | Not provided | Closed Source |
Weights Open Source | Not provided | Not provided |
Commercial use | 不可以商用 | 免费商用授权 |
ResourcesPaper / report | MiniMax M2.7: Early Echoes of Self-Evolution | MiniMax M2.5: Built for Real-World Productivity. |
DataLearner blog | MiniMax M2.7 发布:模型开始帮自己训练自己 | Not provided |

MiniMax M2.5
MiniMaxAI