BAGEL-7B-MoT model card (Hugging Face) May 20, 2025
- MME (sum) 2388
- MMBench 85
- MMMU 55.3
- MM-Vet 67.2
- MathVista 73.1
- GenEval Overall 88
- GEdit-Bench-EN (O) 6.52, SC 7.36, PQ 6.83 6.52
- IntelligentBench 44.0 (BAGEL+CoT 55.3) 55.3
BAGEL-7B-MoT
BAGEL-7B-MoT places in the top three on MME.
17
Benchmarks with a score
0
Ranked first, out of five or more models
1
In the top three
ByteDance Seed May 2025 BAGEL-7B-MoT
Open weights Small
No Epoch Capabilities Index score yet
- PaintBench 5.00% #4 /7
- MathVista 73.1% #41 /70
- MMVP 69.3% #4 /5
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.
Agents and tools
| Benchmark | Score | Rank | Source | |
|---|---|---|---|---|
| GameHorizon | 61.4 | #30 of 37 | Official leaderboard |
Vision, audio and robotics
| Benchmark | Score | Rank | Source | |
|---|---|---|---|---|
| MME | 2388.0 | #3 of 6 | Lab model card | |
| VDEBench | 0.71 | #2 of 4 | Official leaderboard | |
| PaintBench | 5.00% | #4 of 7 | Official leaderboard | |
| MathVista | 73.1% | #41 of 70 | Lab model card | |
| MM-Vet | 67.2% | #5 of 8 | Lab model card | |
| MMBench | 85% | #7 of 11 | Lab model card | |
| GenEval | 88% | #2 of 3 | Lab model card | |
| MMVP | 69.3% | #4 of 5 | Lab model card | |
| MMMU | 55.3% | #91 of 103 | Lab model card | |
| VBVR-Bench | 0.09 | #8 of 8 | Official leaderboard | |
| VBVR-Pro | 0.09 | #8 of 8 | Official leaderboard | |
| GEdit-Bench | 6.52 | #1 of 1 | Lab model card | |
| IntelligentBench | 55.3 | #1 of 1 | Lab model card | |
| KRIS-Bench | 60.2 | #1 of 1 | Lab model card | |
| RISEBench | 11.9% | #1 of 1 | Lab model card | |
| WISE | 0.70 | #1 of 1 | Lab model card |
Its launch post
The benchmarks the lab chose to headline, with the scores it gave.
Its system card
Every result in the longer documents the lab published with it. 4 of them aren't in the launch post.
Capabilities
- MME 2388
- MMBench 85
- MMMU 55.3
- MM-Vet 67.2
- MathVista 73.1
- MMVP 69.3
- GenEval 88
- WISE 0.7
- GEdit-Bench 6.52
- IntelligentBench 55.3
- RISEBench 11.9
- KRIS-Bench 60.2