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 M3
MiniMaxAI
Best overall
MiniMax M3 · 76.76
Best single
MiniMax M3 · BrowseComp 83.50
Modality coverage
MiniMax M3 · 3 modalities
Head to head
2
Benchmarks
2
Wins
0
Losses
+8.54
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.
2 benchmarks with comparable scores. Each model shows its best score; mode label is displayed below.
| Benchmark | MiniMax M3 | MiniMax M2.5 |
|---|---|---|
BrowseComp AI Agent - 信息收集 | 83.50Thinking Enabled ｜ Tools | 76.30Thinking Enabled ｜ Tools |
LiveBench 综合评估 | 70.02Deep Thinking Mode | 60.14Deep Thinking Mode |
Side-by-side input/output token pricing
Licensing, MoE architecture, and multi-modality support.
| Features & specs | MiniMax M3MiniMaxAI | MiniMax M2.5MiniMaxAI |
|---|---|---|
Core specsRelease | 2026-06-01 | 2026-02-12 |
Context length | 1M | 128K |
Parameters | 4280 | 2290 |
Active parameters | 230 | 100 |
Max output | 524288 | Not provided |
MoE | Yes | Yes |
LicenseCode Open Source | Closed Source | Closed Source |
Weights Open Source | Not provided | Not provided |
Commercial use | 不可以商用 | 免费商用授权 |
Modality supportText Input/Output | / | Not provided |
Image Input/Output | / | Not provided |
Video Input/Output | / | Not provided |
ResourcesPaper / report | MiniMax M3: Coding & Agentic Frontier with MSA Architecture, 1M Context and Native Multimodality | MiniMax M2.5: Built for Real-World Productivity. |
MiniMax M2.5
MiniMaxAI
