- SWE-Bench Verified 80.5
- SWE-Bench Pro 59
- Terminal Bench 2.1 66
- SWE Atlas-QnA 37.9
- nl2repo 42.1
- SWE Atlas-Test Writing 30.8
- SWE-fficiency 34.8
- LiveSQLBench 40.2
- CL-bench 20.5
- VIBE-V2 50.1
- SVG-Bench 63.7
- PostTrainBench 37.1
- KernelBench Hard 28.8
- PaperBench 52.6
- BrowseComp 83.5
- DRACO 73.2
- GDPval rubrics 74.8
- BankerToolBench 76.1
- OfficeQA Pro 45.1
- SpreadSheetBench-v1 89.3
- YC-Bench 2100000
- LOCA-Bench (256k) 49.3
- MCP Atlas 74.2
- Apex-Agents 27.7
- Claw-Eval 74.5
- OSWorld-Verified 75.2
- OmniDocBench 91.6
- MMMU-Pro 78.1
- Video-MMMU 84.6
- VideoMME (w/ sub) 85.4
- IMO 2025 35
- USAMO 2026 36
MiniMax-M3
MiniMax-M3 ranks first on 2 of the 108 benchmarks it has scores on.
118
Benchmarks with a score
2
Ranked first, out of five or more models
5
In the top three
MiniMax Jun 2026 MiniMax-M3
Open weights Flagship
146.9
#75 of 271
Epoch Capabilities Index
- MMMU-Pro 78.1% #22 /72
- NL2Repo-Bench 42.1% #20 /28
- OSWorld 2.0 4.60% #6 /7
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 | |
|---|---|---|---|---|
| BullshitBench V2 | 64% | #28 of 122 | Official leaderboard | |
| LMArena (Elo) | 1,439 | #78 of 333 | Official leaderboard | |
| BullshitBench | 47.3% | #30 of 107 | Official leaderboard | |
| SEA-HELM | 72.3 | #25 of 62 | Official leaderboard | |
| Wer wird Millionär? LLM Benchmark | 14756.0 | #31 of 76 | Official leaderboard | |
| Extended NYT Connections | 65.1% | #57 of 97 | Official leaderboard | |
| CL-bench | 20.5% | #4 of 6 | Model card | |
| EQ-Bench 4 | 1,140 | #24 of 32 | Official leaderboard | |
| Document Arena | 1,433 | #32 of 37 | Official leaderboard | |
| LOCA-Bench | 49.3% | #2 of 2 | Model card |
Math and science
| Benchmark | Score | Rank | Source | |
|---|---|---|---|---|
| GPQA Diamond | 90.9% | #35 of 219 | Epoch AI | |
| ResearchClawBench | 19.8 | #9 of 23 | Official leaderboard | |
| CritPt | 3.71% | #75 of 160 | Epoch AI | |
| OTIS Mock AIME 2024-2025 | 71.1% | #98 of 202 | Epoch AI | |
| Surface Evolver Bench | 55% | #13 of 26 | Epoch AI | |
| DTBench | 64.9% | #94 of 178 | Official leaderboard | |
| ProofBench | 19% | #32 of 59 | Official leaderboard | |
| SimpleBench | 45.8% | #61 of 100 | Official leaderboard | |
| ForecastBench | 64.2 | #66 of 86 | Official leaderboard | |
| ProofBench v1.1 | 18% | #41 of 51 | Official leaderboard | |
| NatureBench | 6.67% | #16 of 18 | Official leaderboard |
Coding
| Benchmark | Score | Rank | Source | |
|---|---|---|---|---|
| LiveSQLBench | 40.2% | #1 of 36 | Official leaderboard | |
| SWE-bench Verified | 80.5% | #13 of 150 | Model card | |
| PaperBench | 52.6% | #3 of 14 | Model card | |
| EvoCode-Bench v2 | 23.4% | #3 of 13 | Official leaderboard | |
| Monthly-SWEBench | 58% | #4 of 12 | Official leaderboard | |
| SWE-Bench Pro | 59% | #28 of 77 | Model card | |
| OpenHands Index | 57.2% | #13 of 33 | Official leaderboard | |
| WebDev Arena | 1,487 | #46 of 105 | Epoch AI | |
| Code Arena | 1,483 | #59 of 124 | Official leaderboard | |
| SWE-rebench | 47.2% | #7 of 13 | Official leaderboard | |
| Vibe Code Bench | 47.6% | #64 of 115 | Official leaderboard | |
| PostTrainBench | 37.1% | #4 of 7 | Model card | |
| AndroidBench | 63.6% | #20 of 32 | Official leaderboard | |
| Code Migration | 19.9% | #53 of 78 | Official leaderboard | |
| Senior SWE-Bench | 21.1% | #16 of 23 | Official leaderboard | |
| NL2Repo-Bench | 42.1% | #20 of 28 | Model card | |
| Next.js Agent Evals | 81% | #10 of 14 | Official leaderboard | |
| ALE-Bench | 640 | #89 of 121 | Epoch AI | |
| Image-to-WebDev Arena | 1,470 | #41 of 55 | Official leaderboard | |
| GBAEval | 0.91% | #16 of 21 | Official leaderboard | |
| SWE Atlas | 37.9% | #9 of 11 | Model card | |
| FrontierCode | 14.7% | #41 of 46 | Epoch AI | |
| Code Migration v1.1 | 20.9% | #26 of 29 | Official leaderboard | |
| Vibe Code Bench 1-100 | 9.22% | #24 of 26 | Official leaderboard | |
| Tencent WorkBuddy Bench | 65.5% | #10 of 10 | Official leaderboard | |
| KernelBench | 28.8% | #1 of 1 | Model card | |
| SVG-Bench (MiniMax) | 63.7% | #1 of 1 | Model card | |
| SWE-fficiency | 34.8% | #1 of 1 | Model card | |
| VIBE-V2 (MiniMax) | 50.1% | #1 of 1 | Model card |
Agents and tools
| Benchmark | Score | Rank | Source | |
|---|---|---|---|---|
| Claw-Eval | 74.5% | #3 of 14 | Model card | |
| BrowseComp | 83.5% | #21 of 77 | Model card | |
| OSWorld-Verified | 75.2% | #9 of 33 | Official leaderboard | |
| LiveClawBench | 71.4% | #5 of 17 | Official leaderboard | |
| Pencil Puzzle Bench | 7.10% | #18 of 53 | Official leaderboard | |
| SkillsBench | 53% | #8 of 21 | Official leaderboard | |
| Long-Horizon-Terminal-Bench (LHTB) | 38.5% | #11 of 25 | Official leaderboard | |
| SnakeBench | 25 | #83 of 183 | Official leaderboard | |
| MCP Atlas | 74.2% | #21 of 40 | Model card | |
| Long-Horizon-Terminal-Bench 2.0 | 46.9% | #10 of 15 | Official leaderboard | |
| Harbor-Index | 6.10% | #6 of 9 | Official leaderboard | |
| Terminal-Bench 2.1 | 66% | #37 of 53 | Model card | |
| CivBench | 1,448 | #15 of 18 | Official leaderboard | |
| OSWorld 2.0 | 4.60% | #6 of 7 | Official leaderboard | |
| Chess Puzzles | 13% | #72 of 83 | Epoch AI | |
| Mystery Game Puzzles | 8.00% | #70 of 80 | Epoch AI | |
| Agent Arena | -0.10 | #47 of 53 | Official leaderboard | |
| DRACO | 73.2% | #11 of 11 | Model card |
Professional work
| Benchmark | Score | Rank | Source | |
|---|---|---|---|---|
| CorpFin v2 | 68.1% | #11 of 119 | Official leaderboard | |
| LegalBench | 85.4% | #21 of 138 | Official leaderboard | |
| SpreadsheetBench | 89.3% | #1 of 6 | Model card | |
| MedScribe | 87.3% | #18 of 98 | Official leaderboard | |
| MortgageTax | 68.4% | #16 of 86 | Official leaderboard | |
| MedCode | 46.3% | #36 of 96 | Official leaderboard | |
| FORTE | 71.2% | #8 of 21 | Official leaderboard | |
| TaxEval v2 | 72.7% | #58 of 129 | Official leaderboard | |
| Public Benefits Bench | 64.1% | #24 of 50 | Official leaderboard | |
| Harvey LAB (Legal Agent Benchmark) | 4.17% | #39 of 80 | Official leaderboard | |
| Legal Research Bench | 29.8% | #53 of 88 | Official leaderboard | |
| Vals Finance Agent v2 | 48.3% | #51 of 80 | Official leaderboard | |
| Tax Agent Bench | 20.6% | #51 of 79 | Official leaderboard | |
| ProfBench | 44.4% | #48 of 72 | Official leaderboard | |
| RadLE 2.0 | 350.0 | #8 of 12 | Official leaderboard | |
| EnterpriseClawBench | 52.5% | #6 of 9 | Official leaderboard | |
| Vending-Bench 2 | $2,158 | #49 of 66 | Official leaderboard | |
| Excel Modeling Benchmark | 47.8% | #57 of 76 | Official leaderboard | |
| APEX-Agents | 37.7% | #45 of 56 | Epoch AI | |
| Vals Finance Index | 48.4% | #30 of 35 | Official leaderboard | |
| PolicyBench | 75.4% | #44 of 50 | Official leaderboard | |
| Vals Finance Agent v2.1 | 54.3% | #31 of 35 | Official leaderboard | |
| Excel Modeling Benchmark v1.1 | 59.8% | #33 of 37 | Official leaderboard | |
| Clipping Bench | $223 | #18 of 19 | Official leaderboard | |
| Immigration Bench | 57.5% | #23 of 24 | Official leaderboard | |
| Thrive Holdings helpdesk ticket resolution benchmark | 23.1% | #15 of 15 | Official leaderboard | |
| BlenderWorld | 31.8 | #13 of 13 | Official leaderboard | |
| OfficeQA | 45.1% | #12 of 12 | Model card | |
| BankerToolBench | 76.1% | #3 of 3 | Model card | |
| GDPval-Rubrics (MiniMax) | 74.8% | #1 of 1 | Model card | |
| SpreadsheetBench Verified | 82.5% | #1 of 1 | Official leaderboard | |
| YC-Bench | $2,100,000 | #1 of 1 | Model card |
Vision, audio and robotics
| Benchmark | Score | Rank | Source | |
|---|---|---|---|---|
| MMMU-Pro | 78.1% | #22 of 72 | Model card | |
| Video-MMMU | 84.6% | #11 of 30 | Model card | |
| Video-MME | 85.4% | #16 of 42 | Model card | |
| Vision Arena | 1,233 | #60 of 136 | Official leaderboard | |
| VoxelBench | 1,365 | #47 of 56 | Official leaderboard | |
| OmniDocBench | 91.6% | #3 of 3 | Model card |
Safety and security
| Benchmark | Score | Rank | Source | |
|---|---|---|---|---|
| CyberBench (Vals) | 76.3% | #8 of 26 | Official leaderboard | |
| CyberGym | 73.1% | #14 of 33 | Official leaderboard | |
| CyberBench v1.1 (Vals) | 65% | #29 of 43 | Official leaderboard | |
| DistressBench | 67.7% | #32 of 42 | Official leaderboard | |
| DrugDiscoveryBench | 22.8% | #19 of 19 | Official leaderboard |
Other
| Benchmark | Score | Rank | Source | |
|---|---|---|---|---|
| SAGE (Vals) | 50.6% | #16 of 81 | Official leaderboard | |
| ACCoRD | 65.9 | #67 of 150 | Official leaderboard | |
| SAGE v1.1 (Vals) | 85.5% | #15 of 28 | Official leaderboard | |
| Conceptual Reasoning Index | 46.4 | #80 of 149 | Official leaderboard | |
| LMCA | 33.7% | #75 of 139 | Epoch AI | |
| Vals Multimodal Index | 60% | #18 of 33 | Official leaderboard | |
| SuperCLUE | 56.9 | #14 of 18 | Official leaderboard |
MiniMax's models over time
MiniMax's models in color, against every other model on the Epoch Capabilities Index, by release date.
Its launch post
The benchmarks the lab chose to headline, with the scores it gave.