- DeepSWE v1.1 71.9
- ProgramBench 26.5
- MiMo Code Bench 63.2
- AutomationBench v1.0.6 53.1
- Toolathlon-Verified 76.9
- GDPval-AA 2.1 1673
- Agents Last Exam 31.6
- Terminal Bench 4.0 34.9
- Terminal Bench 2.1 89.9
- OSWorld-Verified 82
- JobBench 62
- CyberGym 94
- MiMo Cyber Bench 80.2
- ExploitGym 17.8
- ExploitBench 47.9
- SEC Bench Pro 66.3
- MiMo Visual Coding 72.3
MiMo-V2.6 Pro
MiMo-V2.6 Pro places in the top three on CyberGym.
Xiaomi Corp Released 21 Sep 2026
55
Benchmarks with a score
0
Ranked first, out of five or more models
1
In the top three
Xiaomi Corp Sep 2026
MiMo-V2.6 Pro
No Epoch Capabilities Index score yet
- Harvey LAB (Legal Agent Benchmark) 10.8% #17 /80
- CritPt 26.6% #18 /160
- APEX-Agents 54.4% #24 /56
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) | 1,480 | #26 of 333 | Official leaderboard |
Math and science
| Benchmark | Score | Rank | Source | |
|---|---|---|---|---|
| CritPt | 26.6% | #18 of 160 | Epoch AI | |
| ProofBench | 70% | #18 of 83 | Epoch AI | |
| ProofBench v1.1 | 70% | #19 of 51 | Official leaderboard | |
| FrontierPhysics | 1.20% | #15 of 19 | Official leaderboard | |
| MysteryMechanism | 15.3% | #27 of 29 | Official leaderboard |
Coding
| Benchmark | Score | Rank | Source | |
|---|---|---|---|---|
| WebDev Arena | 1,618 | #14 of 105 | Epoch AI | |
| Vibe Code Bench | 85.2% | #16 of 115 | Official leaderboard | |
| Code Arena | 1,625 | #22 of 124 | Official leaderboard | |
| Image-to-WebDev Arena | 1,598 | #16 of 55 | Official leaderboard | |
| DeepSWE | 71.9% | #15 of 51 | Model card | |
| Code Migration | 43% | #27 of 78 | Official leaderboard | |
| ALE-Bench | 1,158 | #47 of 121 | Epoch AI | |
| Code Migration v1.1 | 48.8% | #16 of 29 | Official leaderboard | |
| IOI v2 (Vals) | 39.3% | #36 of 45 | Official leaderboard | |
| ProgramBench | 26.5% | #17 of 20 | Model card | |
| MiMo Code Bench | 63.2% | #1 of 1 | Model card | |
| MiMo Visual Coding | 72.3% | #1 of 1 | Model card |
Agents and tools
| Benchmark | Score | Rank | Source | |
|---|---|---|---|---|
| Terminal-Bench 2.1 | 89.9% | #4 of 53 | Model card | |
| AutomationBench | 53.1% | #7 of 41 | Model card | |
| OSWorld-Verified | 82% | #7 of 37 | Model card | |
| Agent Arena | 0.03 | #17 of 53 | Official leaderboard | |
| Toolathlon-Verified | 76.9% | #5 of 13 | Model card | |
| Terminal-Bench 4.0 | 34.9% | #17 of 23 | Model card |
Professional work
| Benchmark | Score | Rank | Source | |
|---|---|---|---|---|
| MedScribe | 88.3% | #13 of 98 | Official leaderboard | |
| Public Benefits Bench | 68.9% | #8 of 50 | Official leaderboard | |
| Legal Research Bench | 47.1% | #16 of 88 | Official leaderboard | |
| Vals Finance Agent v2 | 57.3% | #15 of 80 | Official leaderboard | |
| Harvey LAB (Legal Agent Benchmark) | 10.8% | #17 of 80 | Official leaderboard | |
| Tax Agent Bench | 32.4% | #19 of 79 | Official leaderboard | |
| Vals Legal Index | 72.1% | #9 of 33 | Official leaderboard | |
| Vals Finance Agent v2.1 | 64.3% | #11 of 35 | Official leaderboard | |
| JobBench | 62% | #4 of 12 | Model card | |
| Harvey LAB v1.1 (HLAB-Vals-1.1) | 13.8% | #12 of 33 | Official leaderboard | |
| Vals Finance Index | 57.9% | #13 of 35 | Official leaderboard | |
| MedCode | 45% | #37 of 96 | Official leaderboard | |
| Excel Modeling Benchmark | 62.9% | #30 of 76 | Official leaderboard | |
| APEX-Agents | 54.4% | #24 of 56 | Epoch AI | |
| Agents' Last Exam | 31.6% | #11 of 25 | Model card | |
| Vals Index v2.2 | 57.8% | #16 of 32 | Official leaderboard | |
| Immigration Bench | 67.5% | #12 of 24 | Official leaderboard | |
| GDPval-AA v2.1 | 1,673 | #4 of 8 | Model card | |
| Excel Modeling Benchmark v1.1 | 67.3% | #26 of 37 | Official leaderboard |
Vision, audio and robotics
| Benchmark | Score | Rank | Source | |
|---|---|---|---|---|
| Vision Arena | 1,245 | #50 of 136 | Official leaderboard | |
| Humanity's Sixth Sense | 25.2% | #23 of 29 | Official leaderboard |
Safety and security
| Benchmark | Score | Rank | Source | |
|---|---|---|---|---|
| CyberGym | 94% | #2 of 33 | Model card | |
| CyberBench v1.1 (Vals) | 72.9% | #13 of 43 | Official leaderboard | |
| SEC-bench Pro | 66.3% | #6 of 10 | Model card | |
| SRE-Bench | 3.05% | #20 of 33 | Official leaderboard | |
| ExploitGym | 17.8% | #10 of 14 | Model card | |
| ExploitBench | 47.9% | #14 of 19 | Model card | |
| MiMo Cyber Bench | 80.2% | #1 of 1 | Model card |
Other
| Benchmark | Score | Rank | Source | |
|---|---|---|---|---|
| LiveBench | 68.5% | #13 of 28 | Official leaderboard | |
| SAGE (Vals) | 45% | #41 of 81 | Official leaderboard | |
| SAGE v1.1 (Vals) | 84.5% | #19 of 28 | Official leaderboard |
Its launch post
The benchmarks the lab chose to headline, with the scores it gave.
Its system card
Every result in the longer documents the lab published with it.
MiMo-V2.6 Technical Report 22 Sep 2026
Capabilities
- DeepSWE 71.9
- ProgramBench 26.5
- MiMo Code Bench 63.2
- AutomationBench 53.1
- Toolathlon-Verified 76.9
- GDPval-AA v2.1 1673
- Agents' Last Exam 31.6
- Terminal-Bench 4.0 34.9
- Terminal-Bench 2.1 89.9
- OSWorld-Verified 82
- JobBench 62
- MiMo Visual Coding 72.3
Cyber risk
- CyberGym 94
- MiMo Cyber Bench 80.2
- ExploitGym 17.8
- ExploitBench 47.9
- SEC-bench Pro 66.3