Vision Model Leaderboard: Best Vision-Language Models
Ranked by LMArena Vision Arena scores: users upload an image, ask a question, and vote between two anonymous models’ answers.
- Top open-weightGemma 4 31B1285Google
In LMArena Vision Arena data published Oct 19, 2026, Anthropic’s Claude Fable 5 (High) ranks #1 of 167 models with 1325 (14,714 votes).
Full leaderboard · Creative writing
26 modelsSorted by LMArena score by default; ± is half the 95% confidence interval. Category columns show the model’s score on that sub-board; – means no score.
| # | Model | LMArena · Creative writing | License | ||
|---|---|---|---|---|---|
| Score↓ | Rank | Votes | |||
| 1 | Gemma 4 31BGoogleOpen |
1285±12 | 28 | 2,703 | Apache 2.0 |
| 2 |  DeepSeek-V4.1-Flash-20260910 (Max)DeepSeekOpen |
1275±33 | 34 | 345 | MIT |
| 3 |  Kimi K2.6Moonshot AIOpen |
1273±20 | 35 | 916 | Modified MIT |
| 4 |  GLM-5.3-FlashZ.aiOpen |
1270±17 | 39 | 1,355 | MIT |
| 5 | Kimi K2.5 (Thinking)Moonshot AIOpen |
1269±14 | 40 | 1,768 | Modified MIT |
| 6 |  Qwen3.5-397B-A17BAlibabaOpen |
1260±13 | 44 | 2,304 | Apache 2.0 |
| 7 | Gemma 4 26B A4BGoogleOpen |
1256±16 | 50 | 1,495 | Apache 2.0 |
| 8 | Qwen3.8-27BAlibabaOpen |
1237±23 | 60 | 696 | Apache 2.0 |
| 9 | Kimi K2.5 (Instant)Moonshot AIOpen |
1237±35 | 61 | 286 | Modified MIT |
| 10 |  MiniMax-M3MiniMaxOpen |
1234±17 | 64 | 1,339 | MiniMax Community License |
| 11 |  MiMo-V2.6-FlashXiaomiOpen |
1231±36 | 67 | 303 | MIT |
| 12 | NNVIDIA-Nemotron-3.5-Super-VL-120B-A12B (NVFP4)NVIDIAOpen | 1231±44 | 68 | 193 | OpenMDW-1.1 |
| 13 | MiMo-V2.5XiaomiOpen |
1230±14 | 70 | 1,908 | MIT |
| 14 | Qwen3.5-27BAlibabaOpen |
1224±14 | 72 | 2,140 | Apache 2.0 |
| 15 | Qwen3.5-122B-A10BAlibabaOpen |
1220±21 | 74 | 865 | Apache 2.0 |
| 16 | Qwen3-VL-235B-A22B-InstructAlibabaOpen |
1220±20 | 75 | 974 | Apache 2.0 |
| 17 |  Mistral Large 3Mistral AIOpen |
1217±21 | 77 | 879 | Apache 2.0 |
| 18 | TInkling-SmallThinking MachinesOpen | 1205±24 | 83 | 675 | Apache 2.0 |
| 19 | Qwen3-VL-235B-A22B-ThinkingAlibabaOpen |
1203±44 | 84 | 178 | Apache 2.0 |
| 20 | Mistral Medium 3.5Mistral AIOpen |
1198±33 | 87 | 343 | Modified MIT |
| 21 | Mistral Small 3.2 (2506)Mistral AIOpen |
1181±29 | 93 | 466 | Apache 2.0 |
| 22 | Gemma 3 27B ITGoogleOpen |
1177±25 | 95 | 627 | Gemma |
| 23 |  Llama 4 Scout 17B 16E InstructMetaOpen |
1126±41 | 101 | 238 | Llama |
| 24 | Mistral Small 3.1 24B Instruct (2503)Mistral AIOpen |
1123±24 | 102 | 771 | Apache 2.0 |
| 25 | GLM-4.6VZ.aiOpen |
1113±41 | 103 | 241 | MIT |
| 26 | Llama 4 Maverick 17B 128E InstructMetaOpen |
1099±43 | 104 | 214 | Llama 4 |
Scores: LMArena leaderboard-dataset (lmarena-ai on Hugging Face, CC-BY-4.0, vision config), Creative writing published Oct 19, 2026. Please credit the source when citing.
Sources & methodology
- LMLMArena leaderboard dataset (Hugging Face, CC-BY-4.0)
Vision config, published Oct 19, 2026; 167 models on the overall board.
- Scores are Bradley-Terry Arena ratings fitted from human votes between two anonymous answers to the same image prompt.
- Categories (OCR, diagrams, homework, captioning, Chinese/English…) are separate LMArena boards computed on prompts of that type, not a breakdown of the overall score; ranks can differ a lot.
- Fields without public data (price, context length, release date…) are not shown; blanks are –. Nothing is estimated.
- Synced daily. JSON API: /api/public/vision-leaderboard.php
This week
Collecting data: 1/7 days recorded; weekly changes appear after 7 days
Synced automatically once a day; ranks are compared only between real snapshots, never backfilled.FAQ
What is the best vision-language model right now?
In LMArena Vision Arena data published Oct 19, 2026, Anthropic’s Claude Fable 5 (High) ranks #1 of 167 models with 1325 (14,714 votes).
What is each vendor’s best vision model?
Top overall entry per vendor: Anthropic Claude Fable 5 (High) (1325, #1), Google Gemini 3.8 Flash (High) (1315, #7), Alibaba Qwen3.8-Max (1313, #10), Meta Muse Spark 1.3 (Max) (1308, #12), OpenAI GPT-5.4 (High) (1303, #18).
Which model is best at OCR?
On LMArena’s OCR board (123 models), Claude Fable 5 (High) is #1 with 1337.
What is the best open-weight vision model?
50 open-weight models are listed; the top one is GLM-5.3-Flash (1294, MIT).
How is this different from benchmarks like MMMU?
MMMU and similar benchmarks auto-score a fixed question set; this board comes from blind human votes on people’s own images and questions, reflecting real-use preference. They are not convertible.