- PinchBench (avg.) 81
- ClawEval
- GDPVal-AA 1426
- SWE-bench Verified 78
- SWE-bench Multilingual 71.7
- Terminal-Bench 2.0 57.1
- τ2-bench (Telecom) 96.8
- DeepSearch QA-F1 86.7
MiMo-V2-Pro
MiMo-V2-Pro.
26
Benchmarks with a score
0
Ranked first, out of five or more models
0
In the top three
Xiaomi Corp Mar 2026
MiMo-V2-Pro
Closed weights
151.3
#86 of 404
Self-Reported Index, from the lab's own numbers. No Epoch Capabilities Index score yet.
- Claw-Eval 77% #12 /24
- CL-bench 15.7% #19 /24
- QwenClawBench 56.5% #6 /12
Where it ranks
Its best score on each benchmark, and where that ranks among every model with a score there. We rank each model once, at its best setting.
Knowledge and language
| Benchmark | Score | Rank | Source | |
|---|---|---|---|---|
| LMArena (Elo) | 1,448 | #66 of 335 | Official leaderboard | |
| Longform Creative Writing (EQ-Bench) | 70.9 | #40 of 119 | Official leaderboard | |
| OpenAI MRCR v2 | 23.5% | #37 of 47 | Official leaderboard | |
| CL-bench | 15.7% | #19 of 24 | Official leaderboard | |
| Extended NYT Connections | 25.8% | #85 of 103 | Official leaderboard | |
| CL-bench Life | 6.90% | #15 of 16 | Official leaderboard |
Math and science
| Benchmark | Score | Rank | Source | |
|---|---|---|---|---|
| ResearchClawBench | 15.3 | #17 of 23 | Official leaderboard |
Coding
| Benchmark | Score | Rank | Source | |
|---|---|---|---|---|
| SWE-bench Verified | 78% | #27 of 155 | Lab model card | |
| ALE-Bench | 1,347 | #70 of 126 | Official leaderboard | |
| SWE-bench Multilingual | 71.7% | #42 of 73 | Lab model card | |
| WebDev Arena | 1,433 | #61 of 105 | arena.ai, via Epoch AI | |
| Code Arena | 1,434 | #78 of 128 | Official leaderboard | |
| MiMo Coding Bench | 71.5% | #4 of 5 | Official leaderboard |
Agents and tools
| Benchmark | Score | Rank | Source | |
|---|---|---|---|---|
| tau2-bench | 96.8% | #4 of 79 | Lab model card | |
| Terminal-Bench 2.0 | 57.1% | #26 of 63 | Lab model card | |
| WildClawBench | 48.1% | #15 of 31 | Official leaderboard | |
| Claw-Eval | 77% | #12 of 24 | Official leaderboard | |
| QwenClawBench | 56.5% | #6 of 12 | Official leaderboard | |
| DeepSearchQA | 86.7% | #12 of 23 | Lab model card | |
| ClawProBench | 57.9% | #28 of 49 | Official leaderboard | |
| Claw-Eval Pass^3 | 57.8% | #14 of 24 | Official leaderboard | |
| SnakeBench | 21 | #125 of 188 | Official leaderboard | |
| Claw-Eval-Live | 53.3% | #8 of 12 | Official leaderboard | |
| PinchBench | 81% | #12 of 14 | Lab model card | |
| Pencil Puzzle Bench | 0% | #49 of 53 | Official leaderboard |
Professional work
| Benchmark | Score | Rank | Source | |
|---|---|---|---|---|
| GDPval-AA | 1,426 | #11 of 16 | Lab model card |
Its lab's numbers put it at 151.3, number 86 of 404
The Self-Reported Index fits one score to the 6 capability results Xiaomi Corp reported for this model, on the scale of the Epoch Capabilities Index. Labs choose which benchmarks to show and how to run them, so read it as the lab's claim, not an independent measurement. The 90% interval runs from 147.1 to 155.5. How we compute it.
- tau2-bench 96.8
- DeepSearchQA 86.7
- PinchBench 81
- SWE-bench Verified 78
- SWE-bench Multilingual 71.7
- Terminal-Bench 2.0 57.1
Its lab's number next to a leaderboard
Where the benchmark's own leaderboard, or Epoch AI's run, also has this model. Gaps mostly come from different harnesses, effort levels and task subsets. Every pair.
| Benchmark | Lab says | Board says | Gap |
|---|---|---|---|
| PinchBench | 81% PinchBench (avg.) | 78.3% Leaderboard, OpenClaw | +2.7 pts |
Its launch post
The benchmarks the lab chose to headline, with the scores it gave.