Best LLM for understanding images
- Claude Fable 5 leads
- 1325
- models ranked
- 85
- lead over #2
- +4 Elo
LMArena Vision · read October 21
As of October 21, 2026, Anthropic Claude Fable 5 leads 85 models on LMArena Vision with 1325, ahead of Anthropic Claude Opus 5.5 (1321) and Anthropic Claude Opus 5 (1320).
This list narrows the best LLM for general use ranking.
| Rank | Model | LMArena Vision | Evidence | Coding index | Output $/Mtok |
|---|---|---|---|---|---|
| 1 | Claude Fable 5 Anthropic | 1325 | 82.1 | $50.00 | |
| 2 | Claude Opus 5.5 Anthropic | 1321 | 92.2 | $20.00 | |
| 3 | Claude Opus 5 Anthropic | 1320 | 81.8 | $25.00 | |
| 4 | Claude Fable 5.1 Anthropic | 1319 | 89.1 | $50.00 | |
| 5 | Claude Opus 4.7 Anthropic | 1317 | 51.1 | $25.00 | |
| 6 | Claude Opus 4.6 Anthropic | 1316 | 66.9 | $25.00 | |
| 7 | Gemini 3.8 Flash Google | 1315 | 70.5 | $4.125 | |
| 8 | Gemini 3.7 Flash Google | 1314 | 69.9 | $3.75 | |
| 9 | Qwen3.8 Max (0902) Qwen | 1313 | 63.5 | $6.00 | |
| 10 | Muse Spark 1.3 Meta | 1308 | 79.7 | $4.25 | |
| 11 | Gemini 3.5 Flash Google | 1308 | 42.8 | $9.00 | |
| 12 | Muse Spark 1.2 Meta | 1304 | 65.1 | $4.25 | |
| 13 | GPT-5.4 OpenAI | 1303 | 37.7 | $15.00 | |
| 14 | Gemini 3.6 Flash Google | 1299 | 48.3 | $3.75 | |
| 15 | GPT-5.5 OpenAI | 1297 | 67.8 | $30.00 | |
| 16 | Gemini 3.1 Pro Preview Google | 1296 | 35.3 | $12.00 | |
| 17 | Claude Opus 4.8 Anthropic | 1294 | 66.0 | $25.00 | |
| 18 | GLM 5.3 Flash Z.ai | 1294 | 71.0 | $0.50 | |
| 19 | Muse Spark 1.1 Meta | 1293 | 48.9 | $4.25 | |
| 20 | GPT-6.1 Sol OpenAI | 1288 | 88.7 | $10.00 |
How this list is built
Generated from the dataset on every refresh. Nobody picks the order by hand.
Models by their LMArena Vision result, as the board publishes it, best first. Prompts that come with an image. People ask a question, compare two anonymous answers and vote for the better one; the rating is built from those votes, as LMArena publishes it.
- OpenRouter, fetched
- LMArena (Hugging Face dataset), fetched