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。

Kimi K2.5
Moonshot AI
Each axis is a category average, normalized to a 100-point radar.
Relative edge: 综合评估 +6.1 / Relative gap: 生产力知识 -10.0
Relative edge: 生产力知识 +10.0 / Relative gap: 综合评估 -6.1
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 · 66.56
Best single
MiniMax-M2.7 · Claw Bench 91.70
Modality coverage
Kimi K2.5 · 2 modalities
Head to head
8
Benchmarks
3
Wins
5
Losses
-0.43
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.
8 benchmarks with comparable scores. Each model shows its best score; mode label is displayed below.
| Benchmark | Kimi K2.5 | MiniMax-M2.7 |
|---|---|---|
GPQA Diamond 综合评估 | 87.60Thinking Enabled | 87.00Thinking Enabled |
HLE 综合评估 | 50.20Thinking Enabled | Tools | 28.00Thinking Enabled |
LiveBench 综合评估 | 69.07Thinking Enabled | 63.49Deep Thinking Mode |
SWE-Bench Pro - Public 编程与软件工程 | 50.70Thinking Enabled | Tools | 56.20Thinking Enabled | Tools |
GDPval-AA 生产力知识 | 40.00Thinking Enabled | 50.00Thinking Enabled |
AA-LCR 长上下文能力 | 65.00Thinking Enabled | 69.00Thinking Enabled | Tools |
Claw Bench OpenClaw智能体能力综合测评 | 81.70Thinking Enabled | Tools | 91.70Thinking Enabled | Tools |
Pinch Bench OpenClaw智能体能力综合测评 | 84.80Thinking Enabled | Tools | 87.10Thinking Enabled | Tools |
Side-by-side input/output token pricing
Licensing, MoE architecture, and multi-modality support.
| Features & specs | Kimi K2.5Moonshot AI | MiniMax-M2.7MiniMaxAI |
|---|---|---|
Core specsRelease | 2026-01-27 | 2026-03-18 |
Context length | 256K | 200K |
Parameters | 10000 | 2290 |
Active parameters | 320 | 100 |
Max output | 16384 | 204800 |
MoE | Yes | Yes |
Supported modes | 常规模式(Non-Thinking Mode)思考模式(Thinking Mode) | No mode data |
LicenseCode Open Source | Not provided | Not provided |
Weights Open Source | Not provided | Not provided |
Commercial use | 免费商用授权 | 不可以商用 |
Modality supportText Input/Output | / | / |
Image Input/Output | / | Not provided |
ResourcesPaper / report | Kimi K2.5: Visual Agentic Intelligence | MiniMax M2.7: Early Echoes of Self-Evolution |
DataLearner blog | 重磅!Kimi K2.5发布,依然免费开源!原生多模态MoE架构,全球最大规模参数的开源模型之一,官方评测结果比肩诸多闭源模型!可以驱动100个子Agent执行! | MiniMax M2.7 发布:模型开始帮自己训练自己 |

MiniMax-M2.7
MiniMaxAI