GPT-4V
GPT-4V ranks first on 6 of the 26 benchmarks with five or more models scored.
30
Benchmarks with a score
6
Ranked first, out of five or more models
8
In the top three
OpenAI Nov 2023
GPT-4V
Closed weights
No Epoch Capabilities Index score yet
- SEED-Bench 67.3% #2 /9
- A-Bench 70.4% #69 /116
- OlympicArena 33.2% #7 /25
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 | |
|---|---|---|---|---|
| MMBench-CN v1.1 | 72.2% | #86 of 178 | Official leaderboard |
Math and science
| Benchmark | Score | Rank | Source | |
|---|---|---|---|---|
| OlympiadBench | 18% | #1 of 6 | Official leaderboard | |
| OlympicArena | 33.2% | #7 of 25 | Official leaderboard |
Coding
| Benchmark | Score | Rank | Source | |
|---|---|---|---|---|
| Design2Code | 87.2 | #1 of 5 | Official leaderboard |
Agents and tools
| Benchmark | Score | Rank | Source | |
|---|---|---|---|---|
| VisualWebArena | 16.4% | #1 of 7 | Official leaderboard | |
| WebLINX | 10.4 | #1 of 3 | Official leaderboard |
Vision, audio and robotics
| Benchmark | Score | Rank | Source | |
|---|---|---|---|---|
| BLINK | 51.3% | #1 of 11 | Official leaderboard | |
| MathVerse | 39.4% | #1 of 9 | Official leaderboard | |
| DREAM-1K | 34.4% | #1 of 8 | Official leaderboard | |
| CharXiv (DQ) | 79.9% | #13 of 67 | Official leaderboard | |
| MMT-Bench | 61.1% | #3 of 15 | Official leaderboard | |
| SEED-Bench | 67.3% | #2 of 9 | Official leaderboard | |
| CCBench (MMBench Chinese Culture) | 57.3% | #43 of 152 | Official leaderboard | |
| MMBench | 81% | #46 of 155 | Official leaderboard | |
| CharXiv | 37.1% | #28 of 77 | Official leaderboard | |
| We-Math | 31.1% | #9 of 21 | Official leaderboard | |
| MMStar | 56% | #76 of 173 | Official leaderboard | |
| MTVQA-Test | 22% | #8 of 18 | Official leaderboard | |
| MMMU | 56.8% | #67 of 136 | Official leaderboard | |
| MMBench-EN v1.1 | 75.3% | #88 of 178 | Official leaderboard | |
| SAT (Spatial Aptitude Training) | 50.7% | #1 of 2 | Official leaderboard | |
| V* | 55% | #1 of 2 | Official leaderboard | |
| A-Bench | 70.4% | #69 of 116 | Official leaderboard | |
| Video-MME | 63.3% | #16 of 26 | Official leaderboard | |
| MathVista | 49.9% | #27 of 43 | Official leaderboard | |
| MATH-Vision (MathVision) | 22.8% | #84 of 105 | Official leaderboard | |
| OmniBench | 38.2% | #13 of 16 | Official leaderboard | |
| MME | 1409.4 | #17 of 17 | Official leaderboard | |
| MMVP | 38.7% | #1 of 1 | Official leaderboard |
Other
| Benchmark | Score | Rank | Source | |
|---|---|---|---|---|
| OpenVLM Leaderboard | 63.5 | #68 of 173 | Official leaderboard |