Xiaomi
availableShows if the model has enough results for an index.MiMo-V2.6-Pro
MiMo-V2.6-Pro is a reasoning model from Xiaomi in the MiMo-V2.6 family. 19 benchmarks count toward its score, in 4 categories.
IndexOverall score. 50 is the middle.73.5 ±6.7
CoverageShare of the index weight with results.70%
SpeedOutput tokens per second.38/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.N/A
50 is the middle of the board. The range shows the doubt in the index.
CapabilitiesScore per category. 50 is the middle.
50 is the middleResults
19 counted| BenchmarkThe test name. | CategoryThe capability that the test measures. | ResultThe score from the publisher. | IndexThis result on the index scale. | RunThe settings of the run. | DateDate of the result. | Published byThe source of the result. |
|---|---|---|---|---|---|---|
| CyberGym | Agentic | 94.0% | 80.5 | — | — | Zhun Wang et al. |
| Terminal-Bench 2.1 (provider run) | Agentic | 89.9% | 77.1 | — | — | DeepSeek-AI |
| Terminal-Bench 2.1 (provider run) | Agentic | 89.9% | 77.1 | — | — | DeepSeek-AI |
| Artificial Analysis Long Context Reasoning | Reasoning | 86.3% | 67.9 | — | — | Artificial Analysis |
| OSWorld-Verified | Agentic | 82.0% | 70.1 | — | — | Tianbao Xie et al. |
| Toolathlon-Verified | Agentic | 76.9% | 74.7 | — | — | Moonshot AI |
| DeepSWE | Agentic | 71.9% | 76.0 | — | — | Datacurve AI |
| JobBench | Agentic | 62.0% | 76.7 | — | — | Yuetai Li et al. |
| Artificial Analysis SciCode | Coding | 60.9% | 77.4 | — | — | Artificial Analysis |
| GDPval-AA normalized | Agentic | 58.7% | 80.6 | — | — | Artificial Analysis |
| Artificial Analysis AutomationBench | Agentic | 58.6% | 70.1 | — | — | Artificial Analysis |
| AutomationBench | Agentic | 53.1% | 95.0 | — | — | Moonshot AI |
| Artificial Analysis Humanity's Last Exam | Knowledge | 49.4% | 78.1 | — | — | Artificial Analysis |
| Artificial Analysis Intelligence Index | Knowledge | 46.3% | 80.3 | — | — | Artificial Analysis |
| Artificial Analysis Omniscience Accuracy | Knowledge | 34.8% | 56.9 | — | — | Artificial Analysis |
| Agents' Last Exam | Agentic | 31.6% | 71.0 | — | — | DeepSeek-AI |
| Critical Physics Tasks | Reasoning | 26.6% | 94.2 | — | — | Artificial Analysis |
| ProgramBench: Can Language Models Rebuild Programs From Scratch? | Coding | 26.5% | 39.7 | — | — | John Yang et al. |
| Artificial Analysis GDP.pdf | Agentic | 19.2% | 70.3 | — | — | Artificial Analysis |
| ExploitGym | Agentic | 17.8% | 74.0 | — | — | Zhun Wang et al. |
19 benchmarks count, from 20 of 20 results. A grey row does not count. Too few models took that benchmark.