LLaVA-v1.5-13B
LLaVA-v1.5-13B places in the top three on MMDU.
University of Wisconsin-Madison Released October 5th, 2023 Family LLaVA The lab's model card Compare with others 36
Benchmarks with a score
0
Ranked first, out of five or more models
1
In the top three
University of Wisconsin-Madison Oct 2023 LLaVA-v1.5-13B
Open weights Small
No Epoch Capabilities Index score yet
- Video SimpleQA 19.7% #19 /27
- VisualPuzzles 27.6% #26 /29
- SEED-Bench 61.6% #4 /9
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 | |
|---|---|---|---|---|
| Video SimpleQA | 19.7% | #19 of 27 | Official leaderboard | |
| MMBench-CN v1.1 | 55.5% | #156 of 178 | Official leaderboard |
Agents and tools
| Benchmark | Score | Rank | Source | |
|---|---|---|---|---|
| VisualWebBench | 19.4% | #4 of 7 | Official leaderboard |
Vision, audio and robotics
| Benchmark | Score | Rank | Source | |
|---|---|---|---|---|
| SEED-Bench | 61.6% | #4 of 9 | Official leaderboard | |
| CoreCognition | 55.9% | #57 of 119 | Official leaderboard | |
| MMDU | 35.3 | #3 of 5 | Official leaderboard | |
| BLINK | 40.5% | #8 of 11 | Official leaderboard | |
| MME-RealWorld | 28% | #8 of 11 | Official leaderboard | |
| MMBench | 69.2% | #116 of 155 | Official leaderboard | |
| MuirBench | 24.4% | #6 of 8 | Official leaderboard | |
| MMBench-EN v1.1 | 65.8% | #142 of 178 | Official leaderboard | |
| MMT-Bench | 51.7% | #12 of 15 | Official leaderboard | |
| MM-Vet | 35.6% | #136 of 164 | Official leaderboard | |
| Image2Struct | 21.9 | #15 of 18 | Official leaderboard | |
| CSVQA | 7.50% | #10 of 12 | Official leaderboard | |
| A-Bench | 64.8% | #97 of 116 | Official leaderboard | |
| WildVision | 14.4 | #12 of 14 | Official leaderboard | |
| LongVideoBench | 43.1% | #20 of 23 | Official leaderboard | |
| SEED-Bench-2-Plus | 44.6% | #119 of 136 | Official leaderboard | |
| CCBench (MMBench Chinese Culture) | 20.8% | #135 of 152 | Official leaderboard | |
| VisualPuzzles | 27.6% | #26 of 29 | Official leaderboard | |
| MMMU | 36.4% | #122 of 136 | Official leaderboard | |
| WildVision Arena | 891 | #9 of 10 | Official leaderboard | |
| TempCompass | 51.3% | #12 of 13 | Official leaderboard | |
| MathVista | 27.6% | #41 of 43 | Official leaderboard | |
| MMStar | 34.3% | #165 of 173 | Official leaderboard | |
| MATH-Vision (MathVision) | 11.1% | #102 of 105 | Official leaderboard | |
| OCRBench | 33.1% | #11 of 11 | Official leaderboard | |
| TemporalBench | 13.7% | #10 of 10 | Official leaderboard | |
| MathVerse | 7.60% | #9 of 9 | Official leaderboard | |
| DriveBench Behavior | 33.0 | #3 of 3 | Official leaderboard | |
| DriveBench Perception | 23.4 | #3 of 3 | Official leaderboard | |
| DriveBench Planning | 38.9 | #3 of 3 | Official leaderboard | |
| DriveBench Prediction | 37.8 | #3 of 3 | Official leaderboard | |
| SAT (Spatial Aptitude Training) | 41.6% | #2 of 2 | Official leaderboard |
Other
| Benchmark | Score | Rank | Source | |
|---|---|---|---|---|
| OpenVLM Leaderboard | 39.3 | #162 of 173 | Official leaderboard |