GPT-4o Mar 2025
GPT-4o places in the top three on 2 benchmarks.
58
Benchmarks with a score
0
Ranked first, out of five or more models
2
In the top three
OpenAI Mar 2025
GPT-4o
Mar 2025 snapshot Closed weights
No Epoch Capabilities Index score yet
- ZeroBench 1.00% #71 /76
- HiPhO 32.1% #27 /36
- MicroVQA 45.6% #15 /19
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 | |
|---|---|---|---|---|
| Vectara Hallucination Leaderboard (HHEM) | 1.50% | #13 of 148 | Official leaderboard | |
| MMBench-CN v1.1 | 85.2% | #16 of 178 | Official leaderboard | |
| BuzzBench | 69.1 | #2 of 20 | Official leaderboard | |
| EQ-Bench v2 | 83.5 | #10 of 85 | Official leaderboard | |
| ECLeKTic | 29.9% | #7 of 35 | Official leaderboard | |
| LMArena (Elo) | 1,443 | #73 of 335 | Official leaderboard | |
| Phare | 70.7% | #17 of 62 | Official leaderboard | |
| MultiLoKo | 31% | #2 of 7 | Official leaderboard | |
| EQ-Bench 3 | 1,419 | #31 of 78 | Official leaderboard | |
| Wer wird Millionär? LLM Benchmark | 12622.0 | #33 of 77 | Official leaderboard | |
| Creative Writing v3 | 68.8 | #59 of 120 | Official leaderboard | |
| Longform Creative Writing (EQ-Bench) | 60.2 | #61 of 119 | Official leaderboard | |
| FACTS Grounding | 75.9% | #20 of 34 | Official leaderboard | |
| LLM Confabulation Leaderboard | 20.8% | #40 of 58 | Official leaderboard | |
| FACTS Search | 56.8% | #27 of 31 | Official leaderboard |
Math and science
| Benchmark | Score | Rank | Source | |
|---|---|---|---|---|
| HiPhO | 32.1% | #27 of 36 | Official leaderboard | |
| PolyMath | 24.3% | #29 of 38 | Official leaderboard | |
| PHYBench | 6.97% | #27 of 28 | Official leaderboard | |
| AIME 2025 | 11.7% | #52 of 53 | Official leaderboard | |
| HMMT 2025 | 5.83% | #52 of 53 | Official leaderboard | |
| Verifying the Verifiers (Project APE CRED benchmark) | 11.5% | #23 of 23 | Official leaderboard |
Coding
| Benchmark | Score | Rank | Source | |
|---|---|---|---|---|
| Aider Polyglot | 45.3% | #24 of 46 | Official leaderboard | |
| ArtifactsBench (Gemini 2.5 Pro 0506 judge) | 38% | #11 of 18 | Official leaderboard | |
| SWE-bench Multimodal | 30.4% | #6 of 9 | Official leaderboard | |
| BigO(Bench) | 33.1% | #13 of 18 | Official leaderboard | |
| LiveSQLBench | 15.5% | #30 of 36 | Official leaderboard | |
| SWE-bench Verified | 21.6% | #53 of 60 | Official leaderboard | |
| DesignArena Game Dev | 896 | #138 of 143 | Official leaderboard | |
| DesignArena 3D Design | 852 | #132 of 136 | Official leaderboard | |
| DesignArena Website | 824 | #147 of 151 | Official leaderboard | |
| DesignArena UI Component | 876 | #136 of 139 | Official leaderboard | |
| DesignArena Code | 851 | #141 of 143 | Official leaderboard | |
| GSO-Bench | 0% | #30 of 30 | Official leaderboard | |
| ArtifactsBench | 33.5% | #20 of 20 | Official leaderboard | |
| SWE-bench-Live | 11.7% | #12 of 12 | Official leaderboard |
Agents and tools
| Benchmark | Score | Rank | Source | |
|---|---|---|---|---|
| Elimination Game | 5 | #7 of 56 | Official leaderboard | |
| LisanBench | 302.7 | #79 of 133 | Official leaderboard | |
| Pencil Puzzle Bench | 0% | #32 of 53 | Official leaderboard | |
| Step Race | 2 | #36 of 59 | Official leaderboard | |
| PinchBench | 65.4% | #32 of 48 | Official leaderboard | |
| GAIA | 24.3% | #50 of 64 | Official leaderboard | |
| Meta Agents Research Environments (ARE) / Gaia2 | 7.40% | #11 of 13 | Official leaderboard | |
| OSWorld-Verified | 29.3% | #28 of 33 | Official leaderboard | |
| LMArena Search Arena | 1,084 | #37 of 37 | Official leaderboard |
Professional work
| Benchmark | Score | Rank | Source | |
|---|---|---|---|---|
| xbench-Recruitment | 38.9 | #8 of 8 | Official leaderboard | |
| xbench-Marketing | 32.0 | #7 of 7 | Official leaderboard |
Vision, audio and robotics
| Benchmark | Score | Rank | Source | |
|---|---|---|---|---|
| MMBench-EN v1.1 | 86.8% | #8 of 178 | Official leaderboard | |
| Vision Arena | 1,238 | #55 of 136 | Official leaderboard | |
| ERQA | 34.8% | #10 of 17 | Official leaderboard | |
| VisuLogic | 26.3% | #10 of 17 | Official leaderboard | |
| Point-Bench | 29.5% | #13 of 17 | Official leaderboard | |
| MicroVQA | 45.6% | #15 of 19 | Official leaderboard | |
| ZeroBench | 1.00% | #71 of 76 | Official leaderboard | |
| MineBench | 744 | #73 of 75 | Official leaderboard | |
| Blueprint-Bench | 0.13 | #8 of 8 | Official leaderboard |
Safety and security
| Benchmark | Score | Rank | Source | |
|---|---|---|---|---|
| BaxBench | 26.7% | #18 of 34 | Official leaderboard | |
| KORA | 23.6% | #33 of 42 | Official leaderboard | |
| Virology Capabilities Test (VCT) | 18.8% | #7 of 7 | Official leaderboard |