Phi-3 Vision 128K Instruct
MicrosoftOpenMIT
Phi-3 Vision 128K Instruct is a vision-language model from Microsoft. It ranks 167 of 167 on LMArena’s overall vision board with 812±19 from 1,401 votes. License: MIT.
Key figures
- Developer
- Microsoft
- License
- MIT
- First seen (our snapshots)
- 2026-10-26
No data: OCR, Diagrams, Homework, Captioning, Entity recognition, Creative writing, Humor, Chinese.
Who it suits
- Overall #167 of 167: lower half; consider it mainly for cost or deployment reasons.
- Open weights (MIT): you can self-host it.
Neighbours on the overall board
| # | Model | Score | Votes | License | Δ vs this model |
|---|---|---|---|---|---|
| 161 | Nnvila-internal-15b-v1 NVIDIA | 940 | 1,077 | - | +128 |
| 162 | LLLaVA 1.6 34B | 936 | 4,531 | Apache 2.0 | +124 |
| 163 |  CogVLM2 Llama3 Chat 19B Z.ai | 928 | 1,991 | CogVLM2 | +116 |
| 164 | LLLaVA-OneVision Qwen2 72B OV | 923 | 1,321 | Apache 2.0 | +111 |
| 165 | OMiniCPM-V 2.6 OpenBMB | 912 | 1,987 | Apache 2.0 | +100 |
| 166 | MPhi-3.5 Vision Instruct Microsoft | 851 | 2,592 | MIT | +39 |
| 167 | MPhi-3 Vision 128K Instruct Microsoft | 812 | 1,401 | MIT |
Alternatives · More from Microsoft
| # | Model | Score | Votes | License |
|---|---|---|---|---|
| 166 | MPhi-3.5 Vision Instruct Microsoft | 851 | 2,592 | MIT |
Alternatives · Top open-weight models
| # | Model | Score | Votes | License |
|---|---|---|---|---|
| 24 | GLM-5.3-Flash Z.ai | 1294 | 16,254 | MIT |
| 34 |  Kimi K2.6 Moonshot AI | 1282 | 16,368 | Modified MIT |
| 38 |  DeepSeek-V4.1-Flash-20260910 (Max) DeepSeek | 1278 | 4,100 | MIT |
| 39 |  Gemma 4 31B Google | 1276 | 40,219 | Apache 2.0 |
| 44 |  Qwen3.8-27B Alibaba | 1269 | 7,614 | Apache 2.0 |
How to read these scores
- If two scores differ by less than their ± intervals, the models are statistically close.
- Each category board is computed independently; do not compare scores across categories.
- Models with few votes move more; ranks of new entries may change.
FAQ
Where does Phi-3 Vision 128K Instruct rank among vision models?
Overall rank 167 (812); English rank 167 (837) (LMArena, data as of Oct 26, 2026).
How good is Phi-3 Vision 128K Instruct at OCR?
Phi-3 Vision 128K Instruct is not on LMArena’s OCR board yet, so there is no score to cite.
Is Phi-3 Vision 128K Instruct open-weight?
LMArena lists its license as MIT, not proprietary; check the license text for commercial use.
How far is Phi-3 Vision 128K Instruct from #1?
#1 overall is Claude Fable 5 (High) (1325); Phi-3 Vision 128K Instruct has 812, 513 points behind.
Sources
LMArena data published Oct 19, 2026.
Synced here Oct 26, 2026.
Price, context length and release date have no reliable public data here and are not shown; all figures come from snapshots, blanks are –.