LMArena Vision: reading text leaderboard
- Claude Fable 5 leads
- 1337
- models measured
- 76
- lead over #2
- +4 Elo
- median
- 1270
Leader on LMArena Vision: reading text · read October 21
On October 21, 2026, 76 of our catalog's models had a LMArena Vision: reading text result, and Anthropic Claude Fable 5 led with 1337.
Image prompts that need the text in the image read. 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
We collect this benchmark and show it as supporting evidence. It is not part of the Vibe Coding Index, because it measures something other than coding and the index is about coding. Read the methodology
| # | Model | Result | Votes | Interval (±) | Evidence | Coding index | Output $/Mtok |
|---|---|---|---|---|---|---|---|
| 1 | Claude Fable 5Anthropic | 1337 | 10,497 | ±7.3 | 82.1 | $50.00 | |
| 2 | Gemini 3.8 FlashGoogle | 1333 | 3,137 | ±11.2 | 70.5 | $4.125 | |
| 3 | Claude Opus 5Anthropic | 1329 | 12,218 | ±6.9 | 81.8 | $25.00 | |
| 4 | Claude Fable 5.1Anthropic | 1327 | 3,456 | ±10.4 | 89.1 | $50.00 | |
| 5 | Claude Opus 4.7Anthropic | 1327 | 16,736 | ±6.6 | 51.1 | $25.00 | |
| 6 | Claude Opus 4.6Anthropic | 1326 | 19,407 | ±6.3 | 66.9 | $25.00 | |
| 7 | Claude Opus 5.5Anthropic | 1326 | 2,118 | ±13.5 | 92.2 | $20.00 | |
| 8 | Qwen3.8 Max (0902)Qwen | 1322 | 6,896 | ±8.0 | 63.5 | $6.00 | |
| 9 | Gemini 3.7 FlashGoogle | 1322 | 3,653 | ±10.3 | 69.9 | $3.75 | |
| 10 | Muse Spark 1.3Meta | 1319 | 3,701 | ±10.4 | 79.7 | $4.25 | |
| 11 | Gemini 3.5 FlashGoogle | 1317 | 9,712 | ±7.1 | 42.8 | $9.00 | |
| 12 | Muse Spark 1.2Meta | 1313 | 1,343 | ±16.9 | 65.1 | $4.25 | |
| 13 | GPT-5.4OpenAI | 1312 | 21,790 | ±5.6 | 37.7 | $15.00 | |
| 14 | Gemini 3.6 FlashGoogle | 1312 | 6,075 | ±8.4 | 48.3 | $3.75 | |
| 15 | GPT-5.5OpenAI | 1306 | 20,286 | ±5.8 | 67.8 | $30.00 | |
| 16 | Gemini 3.1 Pro PreviewGoogle | 1304 | 32,719 | ±5.0 | 35.3 | $12.00 | |
| 17 | Claude Opus 4.8Anthropic | 1303 | 12,994 | ±6.6 | 66.0 | $25.00 | |
| 18 | Muse Spark 1.1Meta | 1301 | 8,699 | ±7.3 | 48.9 | $4.25 | |
| 19 | GLM 5.3 FlashZ.ai | 1299 | 11,021 | ±6.6 | 71.0 | $0.50 | |
| 20 | Claude Sonnet 5.5Anthropic | 1297 | 1,780 | ±14.4 | 88.9 | $10.00 | |
| 21 | Claude Sonnet 4.6Anthropic | 1296 | 19,839 | ±6.1 | 57.8 | $15.00 | |
| 22 | Grok 4.5xAI | 1293 | 8,562 | ±7.5 | 75.5 | $6.00 | |
| 23 | GPT-6 AstraOpenAI | 1292 | 2,740 | ±11.8 | 88.7 | $50.00 | |
| 24 | Kimi K2.6Moonshot AI | 1291 | 11,758 | ±7.1 | 44.2 | $3.41 | |
| 25 | GPT-5.6 SolOpenAI | 1290 | 7,710 | ±7.8 | 76.7 | $20.00 | |
| 26 | GPT-6.1 SolOpenAI | 1289 | 1,654 | ±14.8 | 88.7 | $10.00 | |
| 27 | Qwen3.7 PlusQwen | 1288 | 9,923 | ±7.0 | Not benchmarked | $1.28 | |
| 28 | Claude Sonnet 5Anthropic | 1288 | 9,827 | ±7.1 | 63.5 | $10.00 | |
| 29 | Mistral Large 4Mistral AI | 1284 | 671 | ±23.1 | 55.3 | $4.18 | |
| 30 | DeepSeek V4.1 FlashDeepSeek | 1282 | 2,786 | ±11.8 | 77.2 | $0.525 | |
| 31 | Step 5 PreviewStepFun | 1282 | 1,102 | ±17.9 | 71.4 | $2.70 | |
| 32 | Gemma 4 31BGoogle | 1282 | 28,054 | ±5.7 | 38.3 | $0.40 | |
| 33 | Qwen3.8 27BQwen | 1281 | 5,124 | ±9.0 | 73.6 | $1.35 | |
| 34 | Gemini 3.5 Flash LiteGoogle | 1280 | 6,262 | ±8.4 | 33.6 | $2.50 | |
| 35 | GPT-5.6 TerraOpenAI | 1278 | 8,104 | ±7.7 | 69.0 | $12.00 | |
| 36 | Kimi K2.5Moonshot AI | 1276 | 19,489 | ±5.7 | 58.5 | $2.50 | |
| 37 | Grok 4.6xAI | 1272 | 4,881 | ±9.2 | 76.4 | $6.00 | |
| 38 | Qwen3.5 397B A17BQwen | 1271 | 22,696 | ±5.2 | 28.6 | $2.34 | |
| 39 | Gemini 2.5 ProGoogle | 1270 | 32,875 | ±4.8 | 37.0 | $10.00 | |
| 40 | GLM 5V TurboZ.ai | 1270 | 30,138 | ±5.2 | 28.8 | $4.00 | |
| 41 | MiMo-V2.6-FlashXiaomi | 1270 | 2,504 | ±12.5 | 61.9 | $0.28 | |
| 42 | MiMo-V2.6-ProXiaomi | 1270 | 1,195 | ±17.3 | 72.6 | $0.87 | |
| 43 | Gemma 4 26B A4BGoogle | 1268 | 16,862 | ±6.5 | 22.8 | $0.255 | |
| 44 | GPT-5.6 LunaOpenAI | 1264 | 7,472 | ±7.9 | 61.5 | $1.20 | |
| 45 | MiniMax M3MiniMax | 1259 | 13,947 | ±6.5 | 57.8 | $1.20 | |
| 46 | GPT-6 SolOpenAI | 1258 | 2,285 | ±12.9 | 82.8 | $10.00 | |
| 47 | GPT-5.2OpenAI | 1257 | 12,905 | ±6.4 | 58.9 | $14.00 | |
| 48 | MiMo-V2.5Xiaomi | 1256 | 20,901 | ±5.8 | 34.0 | $0.238 | |
| 49 | GPT-5.1OpenAI | 1255 | 6,005 | ±8.8 | 53.3 | $10.00 | |
| 50 | GPT-5.4 MiniOpenAI | 1253 | 18,272 | ±6.2 | 28.2 | $4.50 | |
| 51 | Qwen3 VL 235B A22B InstructQwen | 1253 | 7,752 | ±7.8 | Not benchmarked | $1.90 | |
| 52 | Qwen3.5-122B-A10BQwen | 1251 | 9,185 | ±7.2 | 22.9 | $2.08 | |
| 53 | Gemini 3.1 Flash Lite PreviewGoogle | 1246 | 19,353 | ±5.8 | 8.1 | $1.50 | |
| 54 | GPT-5OpenAI | 1246 | 10,674 | ±9.1 | 72.2 | $10.00 | |
| 55 | Qwen3.5-27BQwen | 1246 | 21,340 | ±5.4 | 22.6 | $2.16 | |
| 56 | T Inkling SmallThinking Machines | 1243 | 5,655 | ±8.6 | 29.9 | $1.20 | |
| 57 | Grok 4.7xAI | 1241 | 2,508 | ±12.3 | 75.0 | $6.00 | |
| 58 | Gemini 2.5 FlashGoogle | 1237 | 27,156 | ±4.9 | 28.7 | $2.50 | |
| 59 | Grok 4.3xAI | 1235 | 19,659 | ±5.8 | 22.5 | $2.50 | |
| 60 | Mistral Medium 3.5Mistral AI | 1233 | 3,905 | ±10.4 | 9.2 | $7.50 | |
| 61 | GPT-6 LunaOpenAI | 1225 | 2,489 | ±12.4 | 71.4 | $0.50 | |
| 62 | o3OpenAI | 1218 | 14,380 | ±8.2 | 70.7 | $8.00 | |
| 63 | GPT-4.1OpenAI | 1218 | 11,355 | ±8.9 | 46.5 | $8.00 | |
| 64 | GPT-5.4 NanoOpenAI | 1204 | 18,222 | ±6.3 | Not benchmarked | $1.25 | |
| 65 | GPT-5 MiniOpenAI | 1199 | 7,999 | ±10.3 | 59.8 | $2.00 | |
| 66 | o4 MiniOpenAI | 1197 | 11,774 | ±8.9 | 68.6 | $4.40 | |
| 67 | GPT-4.1 MiniOpenAI | 1188 | 10,464 | ±9.4 | 28.5 | $1.60 | |
| 68 | Mistral Medium 3.1Mistral AI | 1175 | 13,469 | ±7.3 | Not benchmarked | $2.00 | |
| 69 | GLM 4.6VZ.ai | 1171 | 1,484 | ±16.3 | Not benchmarked | $0.90 | |
| 70 | Gemma 3 27BGoogle | 1156 | 6,196 | ±9.9 | 4.9 | $0.35 | |
| 71 | Mistral Medium 3Mistral AI | 1154 | 4,562 | ±10.5 | Not benchmarked | $2.00 | |
| 72 | GPT-5 NanoOpenAI | 1152 | 1,673 | ±15.1 | 34.8 | $0.40 | |
| 73 | GLM 4.5VZ.ai | 1151 | 1,195 | ±16.9 | Not benchmarked | $1.80 | |
| 74 | Llama 4 MaverickMeta Llama | 1143 | 3,083 | ±11.7 | 18.1 | $0.6525 | |
| 75 | Mistral Small 3.1 24BMistral AI | 1132 | 7,209 | ±10.5 | Not benchmarked | $0.555 | |
| 76 | Llama 4 ScoutMeta Llama | 1117 | 2,788 | ±12.4 | 9.1 | $0.30 |
Only models in our catalog are listed, and each appears once with its best published run. 172 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.