Kimi K3 vs MiMo-V2.5-Pro
Kimi K3 
MiMo-V2.5-Pro Verdict
- Kimi K3 leads on quality: 66.5 vs 59.1.
- Kimi K3 is stronger in agents & tools, coding, composite, human preference, knowledge, long context, maths, multimodal.
- MiMo-V2.5-Pro is stronger in instruction following, reasoning.
- MiMo-V2.5-Pro is 9.2× cheaper ($0.653 vs $6.00 per 1M blended).
| Metric | Kimi K3 | MiMo-V2.5-Pro |
|---|---|---|
| BenchLeader Index | 66.5 | 59.1 |
| Agents & tools score | 67.4 | 51.2 |
| Coding score | 65.0 | 58.4 |
| Composite score | 74.2 | 62.4 |
| Human preference score | 71.2 | 61.7 |
| Knowledge score | 68.2 | 57.6 |
| Long context score | 71.1 | 66.4 |
| Maths score | 78.1 | 43.0 |
| Multimodal score | 65.0 | – |
| Instruction following score | – | 77.0 |
| Reasoning score | – | 67.5 |
| Blended price $/M | $6.00 | $0.653 |
| Output speed | 38 tok/s | 36 tok/s |
| Time to first answer | 56.3 s | 60.0 s |
| Context window | 1.0M | 1.0M |
| SciCode | 58.7% | 50.2% |
| APEX-Agents | 39.3% | – |
| FrontierCode | 44.2% | – |
| ProofBench | 87.0% | 22.0% |
| Epoch Capabilities Index | 157.6 | – |
| LMArena Text | – | 1468 |
| LMArena Hard Prompts | – | 1496 |
| LMArena Coding | – | 1521 |
| LMArena WebDev | – | 1475 |
| LMArena Agent | – | -5.1 |
| LiveBench | 79.2% | – |
| LiveBench Reasoning | 90.7% | – |
| LiveBench Coding | 81.5% | – |
| LiveBench Agentic Coding | 62.2% | – |
| LiveBench Mathematics | 84.4% | – |
| LiveBench Data Analysis | 78.7% | – |
| LiveBench Language | 85.5% | – |
| AA Intelligence Index | 43.8 | 26.4 |
| IFBench | – | 79.9% |
| AA-LCR | 88.7% | 79.7% |
| MMMU-Pro | 80.5% | – |
| AA-Omniscience | 19.7 | 3.3 |
| Terminal-Bench Hard | – | 43.2% |
| GPQA Diamond (AA) | 93.5% | 86.6% |
| Humanity's Last Exam (AA) | 46.9% | 35.7% |
| SciCode (AA) | 59.5% | 50.6% |
| τ²-Bench Telecom (AA) | – | 94.2% |
| LiveCodeBench | – | 81.3% |
| MMLU-Pro | – | 84.6% |
| LegalBench | 86.0% | 77.1% |
| CorpFin | 71.6% | 61.4% |
| TaxEval | 75.7% | 73.8% |
| Terminal-Bench 2.1 (Vals) | 80.9% | 57.3% |
| SWE-bench (Vals) | 93.4% | 74.0% |
| GPQA Diamond (Vals) | 92.9% | 82.6% |
| Vals Index | 57.8 | 41.0 |
| EQ-Bench 4 | 1339 | 1208 |
Data as of 2026-09-10. Best configuration of each model; every score links to its source on the model pages.
Kimi K3 vs MiMo-V2.5-Pro: questions
- Is Kimi K3 better than MiMo-V2.5-Pro?
- Kimi K3 leads on quality: 66.5 vs 59.1. The BenchLeader Index combines every independent quality benchmark; Kimi K3 is ahead overall as of 2026-09-10, but check the category scores for your use.
- Is Kimi K3 better than MiMo-V2.5-Pro for coding?
- Kimi K3 scores higher in coding (65 vs 58 on the category index, where 50 is average).
- Is Kimi K3 better than MiMo-V2.5-Pro for agentic tasks?
- Kimi K3 scores higher in agentic tasks (67 vs 51 on the category index, where 50 is average).
- Which is cheaper, Kimi K3 or MiMo-V2.5-Pro?
- MiMo-V2.5-Pro is cheaper: $0.653 against $6.00 per million tokens, blended at three input tokens per output token.
- Which is faster, Kimi K3 or MiMo-V2.5-Pro?
- Kimi K3 streams faster: 38 against 36 output tokens per second.
- Which has the larger context window?
- Both accept 1.0M tokens of context.