InternVL2.5-26B
InternVL2.5-26B places in the top three on MM-IFEval.
Shanghai AI Lab Released December 5th, 2024 Family InternVL2.5 The lab's model card Compare with others 16
Benchmarks with a score
0
Ranked first, out of five or more models
1
In the top three
Shanghai AI Lab Dec 2024 InternVL2.5-26B
Open weights Mid-size
No Epoch Capabilities Index score yet
- HumanEval-V 8.30% #13 /24
- MM-IFEval 48.1% #3 /9
- MM-Vet 65.4% #36 /164
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 | 84.7% | #18 of 178 | Official leaderboard | |
| MM-IFEval | 48.1% | #3 of 9 | Official leaderboard |
Coding
| Benchmark | Score | Rank | Source | |
|---|---|---|---|---|
| HumanEval-V | 8.30% | #13 of 24 | Official leaderboard |
Vision, audio and robotics
| Benchmark | Score | Rank | Source | |
|---|---|---|---|---|
| MMBench-EN v1.1 | 84.4% | #22 of 178 | Official leaderboard | |
| MMStar | 66.5% | #27 of 173 | Official leaderboard | |
| MM-Vet | 65.4% | #36 of 164 | Official leaderboard | |
| CoreCognition | 59.8% | #32 of 119 | Official leaderboard | |
| CharXiv (DQ) | 73.5% | #22 of 67 | Official leaderboard | |
| MMMU | 60% | #55 of 136 | Official leaderboard | |
| CharXiv | 35.9% | #32 of 77 | Official leaderboard | |
| SpaCE-10 | 43.3% | #17 of 36 | Official leaderboard | |
| VSI-Bench-Debiased | 38.4% | #2 of 4 | Official leaderboard | |
| MMSI-Bench | 28% | #17 of 28 | Official leaderboard | |
| MMMU-Pro | 37.1% | #38 of 57 | Official leaderboard | |
| MATH-Vision (MathVision) | 23.1% | #83 of 105 | Official leaderboard |
Other
| Benchmark | Score | Rank | Source | |
|---|---|---|---|---|
| OpenVLM Leaderboard | 71.0 | #30 of 173 | Official leaderboard |