Xiaomi
availableShows if the model has enough results for an index.MiMo-V2.5-Pro
MiMo-V2.5-Pro is a reasoning model from Xiaomi in the MiMo-V2.5 family. 27 benchmarks count toward its score, in 6 categories.
IndexOverall score out of 100.58.1 ±4.6
CoverageShare of the index weight with results.85%
SpeedOutput tokens per second.29/s
Input / 1MUS dollars per 1M input tokens.$0.435
Output / 1MUS dollars per 1M output tokens.$0.87
ContextMaximum tokens in one request.1.05M
EloLMArena rating and rank.1465 (#32)
The index is a score out of 100. The ± range shows how much it can change.
60,919 votes. Elo shows what people prefer. It does not change the score.
CapabilitiesScore per category, out of 100.
Out of 100Results
27 counted| BenchmarkThe test name. | CategoryThe capability that the test measures. | ResultThe score from the publisher. | IndexThis result as a score out of 100. | RunThe settings of the run. | DateDate of the result. | Published byThe source of the result. |
|---|---|---|---|---|---|---|
| τ²-Bench Tool-Agent-User Evaluation | Agentic | 94.2% | 66.4 | — | — | Victor Barres et al. |
| Artificial Analysis GPQA Diamond | Knowledge | 86.6% | 57.3 | — | — | Artificial Analysis |
| MMLU Pro | Knowledge | 84.6% | 53.8 | — | 1 Sept 2026 | Vals AI |
| GPQA Diamond | Knowledge | 82.6% | 54.5 | — | 1 Sept 2026 | Vals AI |
| LiveCodeBench | Coding | 81.4% | 58.3 | — | 1 Sept 2026 | Vals AI |
| Artificial Analysis IFBench | Instruction | 79.9% | 70.2 | — | — | Artificial Analysis |
| Artificial Analysis Long Context Reasoning | Reasoning | 79.7% | 63.4 | — | — | Artificial Analysis |
| SWE-bench | Coding | 74.0% | 56.9 | — | 1 Sept 2026 | Vals AI |
| τ³-Bench Tool-Agent-User Evaluation | Agentic | 72.9% | 57.1 | — | — | Sierra Research |
| Claw-Eval | Agentic | 63.8% | 58.7 | — | — | Bowen Ye et al. |
| Gert Labs Composite Game Benchmark | Agentic | 62.7% | 68.1 | — | — | Gert Labs |
| Artificial Analysis Coding Index | Coding | 60.2% | 61.4 | — | — | Artificial Analysis |
| Terminal-Bench 2.1 | Agentic | 57.3% | 57.9 | — | 21 Sept 2026 | Vals AI |
| SWE-bench Pro | Coding | 57.2% | 59.3 | — | — | Xiang Deng et al. |
| Artificial Analysis SciCode | Coding | 50.6% | 63.1 | — | — | Artificial Analysis |
| Humanity's Last Exam | Knowledge | 48.0% | 69.4 | — | — | Center for AI Safety et al. |
| Artificial Analysis Humanity's Last Exam | Knowledge | 35.7% | 63.3 | — | — | Artificial Analysis |
| Vibe Code Bench v1.1 | Coding | 34.1% | 56.4 | OpenHands | 21 Sept 2026 | Vals AI |
| Humanity's Last Exam without tools | Knowledge | 34.0% | 57.6 | — | — | OpenAI |
| GDPval-AA normalized | Agentic | 30.4% | 58.9 | — | — | Artificial Analysis |
| Artificial Analysis Intelligence Index | Knowledge | 26.0% | 54.9 | — | — | Artificial Analysis |
| Artificial Analysis Agentic Index | Agentic | 22.7% | 55.6 | — | — | Artificial Analysis |
| Artificial Analysis Omniscience Accuracy | Knowledge | 22.4% | 41.6 | — | — | Artificial Analysis |
| ProofBench v1.1 | Math | 22.0% | 58.3 | — | 21 Sept 2026 | Vals AI |
| Code Migration | Coding | 21.6% | 58.7 | — | 21 Sept 2026 | Vals AI |
| Critical Physics Tasks | Reasoning | 4.0% | 46.8 | — | — | Artificial Analysis |
| APEX-Agents-AA | Agentic | 2.4% | 43.0 | — | — | Artificial Analysis / Mercor |
| Agent Arena steerability | Agentic | -4.0 | 62.9 | — | 15 Sept 2026 | LMArena |
| Agent Arena command recovery | Agentic | -4.9 | 61.8 | — | 15 Sept 2026 | LMArena |
| Agent Arena task outcome | Agentic | -9.0 | 57.2 | — | 15 Sept 2026 | LMArena |
27 benchmarks count, from 30 of 30 results. A grey row does not count. Too few models took that benchmark.