LMArena Vision: diagrams leaderboard
- Claude Fable 5 leads
- 1357
- models measured
- 76
- lead over #2
- +9 Elo
- median
- 1281
Leader on LMArena Vision: diagrams · read October 21
On October 21, 2026, 81 of our catalog's models had a LMArena Vision: diagrams result, and Anthropic Claude Fable 5 led with 1357, ahead of Anthropic Claude Opus 5 at 1348.
Image prompts about diagrams. 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.
Role in the index
LMArena Vision: diagrams is shown as published and does not feed any of our indexes. Read the methodology
| # | Model | Result | Votes | Interval (±) | Evidence | Coding index | Output $/Mtok |
|---|---|---|---|---|---|---|---|
| 1 | Claude Fable 5Anthropic | 1357 | 3,956 | ±10.3 | 79.3 | $50.00 | |
| 2 | Claude Opus 5Anthropic | 1348 | 4,520 | ±9.7 | 83.0 | $25.00 | |
| 3 | Claude Opus 5.5Anthropic | 1345 | 711 | ±22.9 | 88.8 | $20.00 | |
| 4 | Gemini 3.8 FlashGoogle | 1345 | 1,159 | ±18.1 | 61.6 | $3.75 | |
| 5 | Claude Fable 5.1Anthropic | 1343 | 1,179 | ±17.7 | 84.2 | $50.00 | |
| 6 | Claude Opus 4.6Anthropic | 1340 | 7,249 | ±8.7 | 57.2 | $25.00 | |
| 7 | Claude Opus 4.7Anthropic | 1336 | 6,240 | ±9.1 | 67.5 | $25.00 | |
| 8 | Muse Spark 1.3Meta | 1330 | 1,352 | ±16.6 | 57.4 | $4.25 | |
| 9 | Gemini 3.7 FlashGoogle | 1329 | 1,364 | ±16.7 | 58.0 | $3.75 | |
| 10 | Claude Sonnet 5.5Anthropic | 1329 | 625 | ±24.3 | 87.3 | $10.00 | |
| 11 | GPT-5.4OpenAI | 1327 | 8,230 | ±8.0 | 64.4 | $15.00 | |
| 12 | GPT-5.5OpenAI | 1325 | 7,798 | ±8.1 | 71.3 | $30.00 | |
| 13 | Gemini 3.5 FlashGoogle | 1323 | 3,618 | ±10.6 | 39.8 | $9.00 | |
| 14 | Qwen3.8 Max (0902)Qwen | 1323 | 2,497 | ±12.2 | 62.4 | $6.00 | |
| 15 | GLM 5.3 FlashZ.ai | 1317 | 3,995 | ±10.1 | 57.7 | $0.50 | |
| 16 | Muse Spark 1.2Meta | 1317 | 544 | ±26.1 | 55.7 | $4.25 | |
| 17 | Claude Opus 4.8Anthropic | 1315 | 4,883 | ±9.5 | 65.0 | $25.00 | |
| 18 | Gemini 3.6 FlashGoogle | 1314 | 2,285 | ±12.9 | 62.5 | $3.75 | |
| 19 | Gemini 3.1 Pro PreviewGoogle | 1311 | 12,416 | ±6.9 | 32.3 | $12.00 | |
| 20 | Step 5 PreviewStepFun | 1308 | 423 | ±29.0 | 61.1 | $2.70 | |
| 21 | Claude Sonnet 4.6Anthropic | 1306 | 7,630 | ±8.4 | 57.3 | $15.00 | |
| 22 | GPT-6.1 SolOpenAI | 1306 | 573 | ±24.9 | 82.4 | $10.00 | |
| 23 | GPT-5.6 SolOpenAI | 1305 | 2,880 | ±11.7 | 73.5 | $20.00 | |
| 24 | Muse Spark 1.1Meta | 1305 | 3,340 | ±10.8 | 70.3 | $4.25 | |
| 25 | Grok 4.5xAI | 1304 | 3,227 | ±11.0 | 58.6 | $6.00 | |
| 26 | Kimi K2.6Moonshot AI | 1298 | 4,304 | ±10.2 | 52.1 | $4.00 | |
| 27 | GPT-6 AstraOpenAI | 1297 | 966 | ±19.5 | 85.1 | $50.00 | |
| 28 | DeepSeek V4.1 FlashDeepSeek | 1297 | 1,085 | ±18.6 | 62.5 | $0.78 | |
| 29 | GPT-5.6 TerraOpenAI | 1295 | 3,090 | ±11.3 | 63.9 | $12.00 | |
| 30 | Claude Sonnet 5Anthropic | 1293 | 3,718 | ±10.4 | 57.7 | $10.00 | |
| 31 | Qwen3.7 PlusQwen | 1292 | 3,752 | ±10.3 | 33.2 | $1.28 | |
| 32 | Mistral Large 4Mistral AI | 1288 | 243 | ±37.3 | 56.4 | $4.18 | |
| 33 | GLM 5V TurboZ.ai | 1286 | 11,284 | ±7.3 | Not benchmarked | $4.00 | |
| 34 | Grok 4.6xAI | 1285 | 1,838 | ±14.2 | 65.7 | $6.00 | |
| 35 | Qwen3.8 27BQwen | 1285 | 1,863 | ±14.2 | 74.2 | $2.55 | |
| 36 | Gemini 3.5 Flash LiteGoogle | 1284 | 2,334 | ±12.9 | 49.0 | $2.50 | |
| 37 | Gemma 4 31BGoogle | 1283 | 10,238 | ±7.7 | 39.0 | $0.40 | |
| 38 | Kimi K2.5Moonshot AI | 1282 | 7,385 | ±7.9 | 38.6 | $2.85 | |
| 39 | MiMo-V2.6-ProXiaomi | 1281 | 473 | ±27.4 | 65.4 | $0.87 | |
| 40 | Qwen3.5 397B A17BQwen | 1277 | 8,623 | ±7.3 | Not benchmarked | $2.34 | |
| 41 | MiMo-V2.6-FlashXiaomi | 1276 | 919 | ±20.0 | 54.7 | $0.28 | |
| 42 | Gemini 2.5 ProGoogle | 1274 | 9,768 | ±7.1 | 19.1 | $10.00 | |
| 43 | GPT-5.6 LunaOpenAI | 1273 | 2,789 | ±11.7 | 58.0 | $1.20 | |
| 44 | GPT-6 SolOpenAI | 1270 | 844 | ±20.9 | 76.0 | $10.00 | |
| 45 | MiMo-V2.5Xiaomi | 1269 | 7,723 | ±8.2 | 50.0 | $0.336 | |
| 46 | MiniMax M3MiniMax | 1266 | 5,289 | ±9.2 | 39.2 | $1.20 | |
| 47 | GPT-5.4 MiniOpenAI | 1266 | 6,940 | ±8.7 | 35.8 | $4.50 | |
| 48 | GPT-5.2OpenAI | 1265 | 4,773 | ±9.5 | 55.9 | $14.00 | |
| 49 | GPT-5.1OpenAI | 1264 | 2,263 | ±12.6 | 40.4 | $10.00 | |
| 50 | Gemma 4 26B A4BGoogle | 1264 | 6,239 | ±8.8 | Not benchmarked | $0.60 | |
| 51 | Qwen3 VL 235B A22B InstructQwen | 1255 | 2,727 | ±11.4 | Not benchmarked | $1.90 | |
| 52 | Grok 4.7xAI | 1254 | 938 | ±19.8 | 66.7 | $6.00 | |
| 53 | Mistral Medium 3.5Mistral AI | 1254 | 1,538 | ±16.0 | 25.5 | $7.50 | |
| 54 | GPT-5OpenAI | 1252 | 2,522 | ±13.3 | 38.8 | $10.00 | |
| 55 | T Inkling SmallThinking Machines | 1251 | 2,071 | ±13.5 | 27.3 | $1.20 | |
| 56 | Qwen3.5-122B-A10BQwen | 1249 | 3,455 | ±10.8 | Not benchmarked | $2.08 | |
| 57 | Qwen3.5-27BQwen | 1249 | 7,927 | ±7.7 | Not benchmarked | $1.56 | |
| 58 | GPT-6 LunaOpenAI | 1245 | 905 | ±20.4 | 57.3 | $0.50 | |
| 59 | Gemini 2.5 FlashGoogle | 1245 | 8,547 | ±7.2 | Not benchmarked | $2.50 | |
| 60 | Gemini 3.1 Flash Lite PreviewGoogle | 1242 | 7,269 | ±8.4 | 22.7 | $1.50 | |
| 61 | Grok 4.3xAI | 1241 | 7,408 | ±8.3 | 37.4 | $2.50 | |
| 62 | o3OpenAI | 1220 | 3,574 | ±12.1 | Not benchmarked | $8.00 | |
| 63 | GPT-4.1OpenAI | 1215 | 2,758 | ±13.0 | Not benchmarked | $8.00 | |
| 64 | GPT-5.4 NanoOpenAI | 1210 | 6,840 | ±8.9 | 44.8 | $1.25 | |
| 65 | GPT-5 MiniOpenAI | 1202 | 1,936 | ±14.8 | 34.2 | $2.00 | |
| 66 | o4 MiniOpenAI | 1197 | 2,826 | ±12.7 | Not benchmarked | $4.40 | |
| 67 | GLM 4.6VZ.ai | 1182 | 578 | ±23.4 | Not benchmarked | $0.90 | |
| 68 | GPT-4.1 MiniOpenAI | 1182 | 2,365 | ±13.7 | Not benchmarked | $1.60 | |
| 69 | Claude Opus 4RetiredAnthropic | 1177 | 461 | ±25.6 | Not benchmarked | $75.00 | |
| 70 | Mistral Medium 3.1Mistral AI | 1175 | 3,741 | ±10.6 | Not benchmarked | $2.00 | |
| 71 | Gemma 3 27BGoogle | 1156 | 1,715 | ±15.1 | Not benchmarked | $0.30 | |
| 72 | Mistral Medium 3Mistral AI | 1155 | 1,423 | ±16.0 | Not benchmarked | $2.00 | |
| 73 | Claude 3.7 SonnetRetiredAnthropic | 1153 | 291 | ±32.3 | Not benchmarked | $15.00 | |
| 74 | GLM 4.5VZ.ai | 1151 | 369 | ±29.8 | Not benchmarked | $1.80 | |
| 75 | Gemini 2.0 FlashRetiredGoogle | 1149 | 1,109 | ±18.3 | Not benchmarked | $0.40 | |
| 76 | GPT-5 NanoOpenAI | 1141 | 518 | ±26.3 | Not benchmarked | $0.40 | |
| 77 | Claude 3.5 SonnetRetiredAnthropic | 1134 | 314 | ±31.7 | Not benchmarked | $15.00 | |
| 78 | Llama 4 MaverickMeta Llama | 1134 | 947 | ±19.3 | Not benchmarked | $1.00 | |
| 79 | Llama 4 ScoutMeta Llama | 1125 | 845 | ±20.1 | Not benchmarked | $0.50 | |
| 80 | Mistral Small 3.1 24BMistral AI | 1119 | 1,627 | ±15.8 | Not benchmarked | $0.555 | |
| 81 | Claude 3.5 HaikuRetiredAnthropic | 1077 | 319 | ±33.0 | Not benchmarked | $4.00 |
Only models in our catalog are listed, and each appears once with its best published run. 184 rows were read from the source on October 21, 2026, and rows that match no catalog model are left out. Every other column is a figure the board publishes for the same run.