Phi-3-vision-128k-instruct
Phi-3-vision-128k-instruct places in the top three on MIRB.
17
Benchmarks with a score
0
Ranked first, out of five or more models
1
In the top three
Microsoft May 2024 Phi-3-vision-128k-instruct
Open weights Small
No Epoch Capabilities Index score yet
- MM-Vet 44.1% #106 /164
- SEED-Bench-2-Plus 64.2% #60 /136
- CoreCognition 55.8% #59 /119
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 | 59.8% | #142 of 178 | Official leaderboard |
Vision, audio and robotics
| Benchmark | Score | Rank | Source | |
|---|---|---|---|---|
| MIRB | 35.2% | #2 of 7 | Official leaderboard | |
| SEED-Bench-2-Plus | 64.2% | #60 of 136 | Official leaderboard | |
| CharXiv (DQ) | 60.5% | #33 of 67 | Official leaderboard | |
| CoreCognition | 55.8% | #59 of 119 | Official leaderboard | |
| CharXiv | 31.6% | #41 of 77 | Official leaderboard | |
| MMBench | 73.6% | #92 of 155 | Official leaderboard | |
| MMBench-EN v1.1 | 70.5% | #114 of 178 | Official leaderboard | |
| MM-Vet | 44.1% | #106 of 164 | Official leaderboard | |
| MMStar | 47.7% | #112 of 173 | Official leaderboard | |
| LongVideoBench | 49.9% | #15 of 23 | Official leaderboard | |
| A-Bench | 68.2% | #85 of 116 | Official leaderboard | |
| MathVista | 44.5% | #33 of 43 | Official leaderboard | |
| CCBench (MMBench Chinese Culture) | 24.1% | #130 of 152 | Official leaderboard | |
| TemporalBench | 15.4% | #9 of 10 | Official leaderboard | |
| Vision Arena | 881 | #136 of 136 | Official leaderboard |
Other
| Benchmark | Score | Rank | Source | |
|---|---|---|---|---|
| OpenVLM Leaderboard | 53.6 | #113 of 173 | Official leaderboard |