LMArena Vision: humor leaderboard
- Claude Fable 5 leads
- 1341
- models measured
- 62
- lead over #2
- +18 Elo
- median
- 1260
Leader on LMArena Vision: humor · read October 21
On October 21, 2026, 62 of our catalog's models had a LMArena Vision: humor result, and Anthropic Claude Fable 5 led with 1341.
Image prompts about what is funny in 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.
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 | 1341 | 583 | ±25.3 | 82.1 | $50.00 | |
| 2 | Gemini 3.1 Pro PreviewGoogle | 1322 | 1,885 | ±14.2 | 35.3 | $12.00 | |
| 3 | Gemini 3.5 FlashGoogle | 1321 | 577 | ±25.4 | 42.8 | $9.00 | |
| 4 | Gemini 3.8 FlashGoogle | 1321 | 175 | ±45.4 | 70.5 | $4.125 | |
| 5 | Claude Fable 5.1Anthropic | 1318 | 208 | ±40.4 | 89.1 | $50.00 | |
| 6 | Claude Opus 4.6Anthropic | 1314 | 1,026 | ±19.4 | 66.9 | $25.00 | |
| 7 | Claude Opus 4.7Anthropic | 1309 | 889 | ±20.9 | 51.1 | $25.00 | |
| 8 | Qwen3.8 Max (0902)Qwen | 1304 | 435 | ±29.2 | 63.5 | $6.00 | |
| 9 | Gemini 3.7 FlashGoogle | 1299 | 215 | ±41.5 | 69.9 | $3.75 | |
| 10 | Claude Opus 5Anthropic | 1293 | 876 | ±20.8 | 81.8 | $25.00 | |
| 11 | Muse Spark 1.3Meta | 1293 | 208 | ±41.7 | 79.7 | $4.25 | |
| 12 | GPT-5.5OpenAI | 1293 | 1,075 | ±18.5 | 67.8 | $30.00 | |
| 13 | Gemini 3.6 FlashGoogle | 1289 | 427 | ±29.3 | 48.3 | $3.75 | |
| 14 | Muse Spark 1.1Meta | 1288 | 577 | ±25.3 | 48.9 | $4.25 | |
| 15 | Gemini 3.5 Flash LiteGoogle | 1287 | 396 | ±30.0 | 33.6 | $2.50 | |
| 16 | GPT-5.4OpenAI | 1284 | 1,143 | ±18.3 | 37.7 | $15.00 | |
| 17 | Kimi K2.5Moonshot AI | 1284 | 1,083 | ±18.1 | 58.5 | $2.50 | |
| 18 | Claude Opus 4.8Anthropic | 1282 | 756 | ±22.2 | 66.0 | $25.00 | |
| 19 | Kimi K2.6Moonshot AI | 1280 | 590 | ±24.5 | 44.2 | $3.41 | |
| 20 | Gemma 4 31BGoogle | 1280 | 1,659 | ±14.8 | 38.3 | $0.40 | |
| 21 | GLM 5.3 FlashZ.ai | 1279 | 671 | ±23.5 | 71.0 | $0.50 | |
| 22 | Grok 4.5xAI | 1276 | 573 | ±25.1 | 75.5 | $6.00 | |
| 23 | Gemma 4 26B A4BGoogle | 1272 | 993 | ±19.2 | 22.8 | $0.255 | |
| 24 | Grok 4.6xAI | 1270 | 320 | ±32.5 | 76.4 | $6.00 | |
| 25 | GLM 5V TurboZ.ai | 1268 | 1,765 | ±14.7 | 28.8 | $4.00 | |
| 26 | Gemini 3.1 Flash Lite PreviewGoogle | 1266 | 979 | ±19.3 | 8.1 | $1.50 | |
| 27 | GPT-5.1OpenAI | 1263 | 376 | ±31.6 | 53.3 | $10.00 | |
| 28 | Gemini 2.5 ProGoogle | 1262 | 3,133 | ±14.9 | 37.0 | $10.00 | |
| 29 | Qwen3.5 397B A17BQwen | 1262 | 1,223 | ±17.2 | 28.6 | $2.34 | |
| 30 | Qwen3.7 PlusQwen | 1261 | 556 | ±25.2 | Not benchmarked | $1.28 | |
| 31 | Claude Sonnet 5Anthropic | 1261 | 628 | ±24.5 | 63.5 | $10.00 | |
| 32 | DeepSeek V4.1 FlashDeepSeek | 1260 | 166 | ±45.5 | 77.2 | $0.525 | |
| 33 | Claude Sonnet 4.6Anthropic | 1248 | 1,060 | ±19.2 | 57.8 | $15.00 | |
| 34 | GPT-5.6 SolOpenAI | 1246 | 569 | ±25.2 | 76.7 | $20.00 | |
| 35 | Qwen3.5-122B-A10BQwen | 1243 | 468 | ±27.3 | 22.9 | $2.08 | |
| 36 | GPT-5.6 TerraOpenAI | 1236 | 526 | ±26.3 | 69.0 | $12.00 | |
| 37 | Qwen3.8 27BQwen | 1235 | 347 | ±32.1 | 73.6 | $1.35 | |
| 38 | GPT-5.4 MiniOpenAI | 1235 | 868 | ±21.1 | 28.2 | $4.50 | |
| 39 | MiniMax M3MiniMax | 1235 | 773 | ±22.1 | 57.8 | $1.20 | |
| 40 | Qwen3.5-27BQwen | 1227 | 1,195 | ±17.6 | 22.6 | $2.16 | |
| 41 | GPT-5.2OpenAI | 1227 | 681 | ±23.6 | 58.9 | $14.00 | |
| 42 | MiMo-V2.5Xiaomi | 1226 | 1,147 | ±18.1 | 34.0 | $0.238 | |
| 43 | Gemini 2.5 FlashGoogle | 1225 | 2,192 | ±15.3 | 28.7 | $2.50 | |
| 44 | o3OpenAI | 1222 | 1,643 | ±19.4 | 70.7 | $8.00 | |
| 45 | Grok 4.3xAI | 1222 | 1,069 | ±18.9 | 22.5 | $2.50 | |
| 46 | GPT-5OpenAI | 1219 | 1,448 | ±20.3 | 72.2 | $10.00 | |
| 47 | Mistral Medium 3.5Mistral AI | 1216 | 238 | ±39.3 | 9.2 | $7.50 | |
| 48 | Qwen3 VL 235B A22B InstructQwen | 1209 | 508 | ±26.3 | Not benchmarked | $1.90 | |
| 49 | T Inkling SmallThinking Machines | 1207 | 384 | ±30.2 | 29.9 | $1.20 | |
| 50 | GPT-5 MiniOpenAI | 1205 | 1,047 | ±22.4 | 59.8 | $2.00 | |
| 51 | o4 MiniOpenAI | 1199 | 1,449 | ±20.1 | 68.6 | $4.40 | |
| 52 | GPT-4.1OpenAI | 1196 | 1,365 | ±20.8 | 46.5 | $8.00 | |
| 53 | GPT-5.6 LunaOpenAI | 1190 | 502 | ±27.4 | 61.5 | $1.20 | |
| 54 | GPT-4.1 MiniOpenAI | 1182 | 1,403 | ±20.8 | 28.5 | $1.60 | |
| 55 | Mistral Medium 3.1Mistral AI | 1168 | 1,502 | ±19.1 | Not benchmarked | $2.00 | |
| 56 | Gemma 3 27BGoogle | 1166 | 762 | ±24.1 | 4.9 | $0.35 | |
| 57 | GPT-5.4 NanoOpenAI | 1161 | 923 | ±21.0 | Not benchmarked | $1.25 | |
| 58 | Mistral Medium 3Mistral AI | 1145 | 440 | ±30.3 | Not benchmarked | $2.00 | |
| 59 | Mistral Small 3.1 24BMistral AI | 1142 | 933 | ±23.3 | Not benchmarked | $0.555 | |
| 60 | GPT-5 NanoOpenAI | 1110 | 174 | ±46.5 | 34.8 | $0.40 | |
| 61 | Llama 4 ScoutMeta Llama | 1102 | 191 | ±44.9 | 9.1 | $0.30 | |
| 62 | Llama 4 MaverickMeta Llama | 1100 | 216 | ±40.4 | 18.1 | $0.6525 |
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.