InternVL2.5-78B-MPO
InternVL2.5-78B-MPO ranks first on CCBench (MMBench Chinese Culture).
Shanghai AI Lab Released December 20th, 2024 Family InternVL2.5 The lab's model card Compare with others 12
Benchmarks with a score
1
Ranked first, out of five or more models
4
In the top three
Shanghai AI Lab Dec 2024 InternVL2.5-78B-MPO
Open weights Flagship
No Epoch Capabilities Index score yet
- MM-Vet 73.5% #14 /164
- SEED-Bench-2-Plus 70.8% #12 /136
- A-Bench 79.3% #6 /116
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 | 87.5% | #4 of 178 | Official leaderboard | |
| SimpleVQA | 45.4% | #3 of 13 | Official leaderboard | |
| Creation-MMBench | -12.55 | #6 of 19 | Official leaderboard |
Vision, audio and robotics
| Benchmark | Score | Rank | Source | |
|---|---|---|---|---|
| CCBench (MMBench Chinese Culture) | 86.9% | #1 of 152 | Official leaderboard | |
| MMBench | 88.4% | #2 of 155 | Official leaderboard | |
| MMBench-EN v1.1 | 87.9% | #3 of 178 | Official leaderboard | |
| MMStar | 72.1% | #6 of 173 | Official leaderboard | |
| A-Bench | 79.3% | #6 of 116 | Official leaderboard | |
| MM-Vet | 73.5% | #14 of 164 | Official leaderboard | |
| SEED-Bench-2-Plus | 70.8% | #12 of 136 | Official leaderboard | |
| MME-CoT | 52.7% | #4 of 10 | Official leaderboard |
Other
| Benchmark | Score | Rank | Source | |
|---|---|---|---|---|
| OpenVLM Leaderboard | 77.0 | #6 of 173 | Official leaderboard |