LLaVA-OneVision 72B
LLaVA-OneVision 72B ranks first on 2 of the 34 benchmarks with five or more models scored.
Released August 6th, 2024 Compare with others
34
Benchmarks with a score
2
Ranked first, out of five or more models
6
In the top three
Unknown lab Aug 2024
LLaVA-OneVision 72B
No Epoch Capabilities Index score yet
- Video SimpleQA 25.5% #11 /27
- VideoReasonBench 0.10% #13 /14
- VisualPuzzles 30.8% #20 /29
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 | 83.9% | #20 of 178 | Official leaderboard | |
| Video SimpleQA | 25.5% | #11 of 27 | Official leaderboard | |
| Creation-MMBench | -39.87 | #16 of 19 | Official leaderboard |
Coding
| Benchmark | Score | Rank | Source | |
|---|---|---|---|---|
| HumanEval-V | 13.8% | #11 of 24 | Official leaderboard |
Vision, audio and robotics
| Benchmark | Score | Rank | Source | |
|---|---|---|---|---|
| MMBench | 85.8% | #14 of 155 | Official leaderboard | |
| TemporalBench | 28.1% | #1 of 10 | Official leaderboard | |
| MMBench-EN v1.1 | 85% | #18 of 178 | Official leaderboard | |
| SpaCE-10 | 52.8% | #4 of 36 | Official leaderboard | |
| VSI-Bench | 40.2% | #1 of 9 | Official leaderboard | |
| MMStar | 65.8% | #28 of 173 | Official leaderboard | |
| CCBench (MMBench Chinese Culture) | 63.9% | #33 of 152 | Official leaderboard | |
| MLVU | 47.2% | #2 of 9 | Official leaderboard | |
| MM-Vet | 60.6% | #52 of 164 | Official leaderboard | |
| VCR-Bench | 36.4% | #3 of 9 | Official leaderboard | |
| CharXiv (DQ) | 71.8% | #24 of 67 | Official leaderboard | |
| DREAM-1K | 33.2% | #3 of 8 | Official leaderboard | |
| CharXiv | 36.4% | #31 of 77 | Official leaderboard | |
| Video-MME | 69.6% | #11 of 26 | Official leaderboard | |
| VLADBench | 50.2% | #3 of 7 | Official leaderboard | |
| MEGA-Bench | 33.8% | #10 of 22 | Official leaderboard | |
| We-Math | 28.7% | #10 of 21 | Official leaderboard | |
| MMMU | 56.8% | #68 of 136 | Official leaderboard | |
| MMSI-Bench | 28.4% | #15 of 28 | Official leaderboard | |
| Video-MMMU | 48.3% | #13 of 24 | Official leaderboard | |
| VisualPuzzles | 30.8% | #20 of 29 | Official leaderboard | |
| MMMU-Pro | 31% | #42 of 57 | Official leaderboard | |
| MATH-Vision (MathVision) | 25.3% | #80 of 105 | Official leaderboard | |
| OST-Bench | 44.6% | #5 of 6 | Official leaderboard | |
| EMMA | 27.3% | #6 of 7 | Official leaderboard | |
| MME-CoT | 36.3% | #9 of 10 | Official leaderboard | |
| VideoReasonBench | 0.10% | #13 of 14 | Official leaderboard | |
| Point-Bench | 18% | #16 of 17 | Official leaderboard |
Safety and security
| Benchmark | Score | Rank | Source | |
|---|---|---|---|---|
| Video-SafetyBench | 3.10 | #14 of 15 | Official leaderboard |
Other
| Benchmark | Score | Rank | Source | |
|---|---|---|---|---|
| OpenVLM Leaderboard | 68.0 | #44 of 173 | Official leaderboard |