InternVL3-78B
InternVL3-78B ranks first on MMBench.
14
Benchmarks with a score
1
Ranked first, out of five or more models
5
In the top three
Shanghai AI Lab Apr 2025 InternVL3-78B
Open weights Flagship
No Epoch Capabilities Index score yet
- HiPhO 37.2% #24 /35
- A-Bench 78.9% #9 /116
- MMMU 72.2% #28 /136
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 | 87.9% | #2 of 178 | Official leaderboard |
Math and science
| Benchmark | Score | Rank | Source | |
|---|---|---|---|---|
| HiPhO | 37.2% | #24 of 35 | Official leaderboard |
Agents and tools
| Benchmark | Score | Rank | Source | |
|---|---|---|---|---|
| SnakeBench | 19 | #136 of 183 | Official leaderboard |
Vision, audio and robotics
| Benchmark | Score | Rank | Source | |
|---|---|---|---|---|
| MMBench | 88.8% | #1 of 155 | Official leaderboard | |
| MMStar | 73.4% | #3 of 173 | Official leaderboard | |
| CCBench (MMBench Chinese Culture) | 83.5% | #3 of 152 | Official leaderboard | |
| MMBench-EN v1.1 | 87.5% | #4 of 178 | Official leaderboard | |
| A-Bench | 78.9% | #9 of 116 | Official leaderboard | |
| CharXiv | 46% | #14 of 77 | Official leaderboard | |
| MMMU | 72.2% | #28 of 136 | Official leaderboard | |
| VisuLogic | 27.7% | #7 of 17 | Official leaderboard | |
| MATH-Vision (MathVision) | 43.1% | #58 of 105 | Official leaderboard | |
| OmniDocBench v1.5 | 80.3 | #10 of 10 | Official leaderboard |
Other
| Benchmark | Score | Rank | Source | |
|---|---|---|---|---|
| OpenVLM Leaderboard | 79.1 | #3 of 173 | Official leaderboard |