Kimi K3 vs MiMo-V2.5
Kimi K3 
MiMo-V2.5 Verdict
- Kimi K3 leads on quality: 66.5 vs 58.4.
- Kimi K3 is stronger in coding, composite, human preference, knowledge, long context, maths, multimodal.
- MiMo-V2.5 is stronger in agents & tools, instruction following.
- MiMo-V2.5 is 34× cheaper ($0.175 vs $6.00 per 1M blended).
- MiMo-V2.5 streams 1.4× faster (52 vs 38 tokens per second).
| Metric | Kimi K3 | MiMo-V2.5 |
|---|---|---|
| BenchLeader Index | 66.5 | 58.4 |
| Agents & tools score | 67.4 | 70.0 |
| Coding score | 65.0 | – |
| Composite score | 74.2 | 57.2 |
| Human preference score | 71.2 | – |
| Knowledge score | 68.2 | 59.5 |
| Long context score | 71.1 | 62.9 |
| Maths score | 78.1 | – |
| Multimodal score | 65.0 | 59.8 |
| Instruction following score | – | 66.1 |
| Blended price $/M | $6.00 | $0.175 |
| Output speed | 38 tok/s | 52 tok/s |
| Time to first answer | 56.3 s | 45.6 s |
| Context window | 1.0M | 1M |
| SciCode | 58.7% | – |
| APEX-Agents | 39.3% | – |
| FrontierCode | 44.2% | – |
| ProofBench | 87.0% | – |
| Epoch Capabilities Index | 157.6 | – |
| 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 | 22.3 |
| IFBench | – | 67.1% |
| AA-LCR | 88.7% | 73.0% |
| MMMU-Pro | 80.5% | 75.4% |
| AA-Omniscience | 19.7 | -9.8 |
| Terminal-Bench Hard | – | 41.7% |
| GPQA Diamond (AA) | 93.5% | 85.0% |
| Humanity's Last Exam (AA) | 46.9% | 27.2% |
| SciCode (AA) | 59.5% | 43.9% |
| τ²-Bench Telecom (AA) | – | 90.6% |
| LegalBench | 86.0% | – |
| CorpFin | 71.6% | – |
| TaxEval | 75.7% | – |
| Terminal-Bench 2.1 (Vals) | 80.9% | – |
| SWE-bench (Vals) | 93.4% | – |
| GPQA Diamond (Vals) | 92.9% | – |
| Vals Index | 57.8 | – |
| EQ-Bench 4 | 1339 | – |
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: questions
- Is Kimi K3 better than MiMo-V2.5?
- Kimi K3 leads on quality: 66.5 vs 58.4. 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 for agentic tasks?
- MiMo-V2.5 scores higher in agentic tasks (70 vs 67 on the category index, where 50 is average).
- Which is cheaper, Kimi K3 or MiMo-V2.5?
- MiMo-V2.5 is cheaper: $0.175 against $6.00 per million tokens, blended at three input tokens per output token.
- Which is faster, Kimi K3 or MiMo-V2.5?
- MiMo-V2.5 streams faster: 52 against 38 output tokens per second.
- Which has the larger context window?
- Kimi K3 accepts more context: 1.0M against 1M tokens.