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 M2.5
MiniMaxAI
Each axis is a category average, normalized to a 100-point radar.
Relative edge: AI Agent - 信息收集 +28.9 / Relative gap: none clear
Relative edge: none clear / Relative gap: AI Agent - 信息收集 -28.9
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 M2.5 · 71.01
Best single
MiniMax M2.5 · τ²-Bench - Telecom 97.80
Modality coverage
M2.1 · 1 modalities
Head to head
10
Benchmarks
8
Wins
1
Losses
+8.21
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.
10 benchmarks with comparable scores. Each model shows its best score; mode label is displayed below.
| Benchmark | MiniMax M2.5 | M2.1 |
|---|---|---|
GPQA Diamond 综合评估 | 85.20Thinking Enabled | 81.00Thinking Enabled |
HLE 综合评估 | 19.40Thinking Enabled | 22.00Thinking Enabled |
SWE-Bench Pro - Public 编程与软件工程 | 55.40Thinking Enabled ｜ Tools | 32.60Thinking Enabled ｜ Tools |
SWE-bench Verified 编程与软件工程 | 80.20Thinking Enabled ｜ Tools | 74.80Thinking Enabled |
AIME2025 数学推理 | 86.30Thinking Enabled | 81.00Thinking Enabled |
τ²-Bench - Telecom Agent能力评测 | 97.80Thinking Enabled ｜ Tools | 87.00Thinking Enabled ｜ Tools |
IF Bench 指令跟随 | 70.00Thinking Enabled ｜ Tools | 70.00Thinking Enabled ｜ Tools |
BrowseComp AI Agent - 信息收集 | 76.30Thinking Enabled ｜ Tools | 47.40Thinking Enabled ｜ Tools |
Terminal Bench 2.0 AI Agent - 工具使用 | 51.70Thinking Enabled ｜ Tools | 47.90Thinking Enabled ｜ Tools |
Pinch Bench OpenClaw智能体能力综合测评 | 87.80Thinking Enabled ｜ Tools | 84.30Thinking Enabled ｜ Tools |
Side-by-side input/output token pricing
Licensing, MoE architecture, and multi-modality support.
| Features & specs | MiniMax M2.5MiniMaxAI | M2.1MiniMaxAI |
|---|---|---|
Core specsRelease | 2026-02-12 | 2025-12-23 |
Context length | 128K | 200K |
Parameters | 2290 | 2300 |
Active parameters | 100 | 100 |
Max output | Not provided | 131072 |
MoE | Yes | Yes |
Supported modes | No mode data | 常规模式（Non-Thinking Mode）思考模式（Thinking Mode） |
LicenseCode Open Source | Closed Source | Closed Source |
Weights Open Source | Not provided | Not provided |
Commercial use | 免费商用授权 | 免费商用授权 |
Modality supportText Input/Output | Not provided | / |
ResourcesPaper / report | MiniMax M2.5: Built for Real-World Productivity. | MiniMax M2.1: Significantly Enhanced Multi-Language Programming, Built for Real-World Complex Tasks |
M2.1
MiniMaxAI
