LMArena Vision Arena
LMArena · vision-2026-10-09-a64ca3b12b6e · Blind human preferences for multimodal responses in LMArena's vision arena.
Model Ranking
Source | |||||||
|---|---|---|---|---|---|---|---|
| #1 | Claude Fable 5Anthropic | 1,323.95 | — | 14,175 | claude-fable-5-high; 95% CI [1316.82313713, 1331.08571856]; votes 14175; rank 1unknown | Original sourcewinner eligible | Third-party benchmark |
| #2 | Claude Opus 5Anthropic | 1,319.22 | — | 16,665 | claude-opus-5-high; 95% CI [1312.47954379, 1325.95953993]; votes 16665; rank 3unknown | Original sourcewinner eligible | Third-party benchmark |
| #3 | Claude Fable 5.1Anthropic | 1,318.24 | — | 4,648 | claude-fable-5.1-max; 95% CI [1308.61778049, 1327.86146111]; votes 4648; rank 4unknown | Original sourcewinner eligible | Third-party benchmark |
| #4 | Claude Opus 4.7Anthropic | 1,316.41 | — | 23,165 | claude-opus-4-7; 95% CI [1309.83870131, 1322.97989133]; votes 23165; rank 5unknown | Original sourcewinner eligible | Third-party benchmark |
| #5 | Gemini 3.7 FlashGoogle DeepMind | 1,315.97 | — | 4,524 | gemini-3.7-flash-high; 95% CI [1306.20108292, 1325.73138242]; votes 4524; rank 6unknown | Original sourcewinner eligible | Third-party benchmark |
| #6 | Claude Opus 4.6Anthropic | 1,315.64 | — | 22,304 | claude-opus-4-6-high; 95% CI [1309.23331588, 1322.04524655]; votes 22304; rank 7unknown | Original sourcewinner eligible | Third-party benchmark |
| #7 | Gemini 3.8 FlashGoogle DeepMind | 1,314.08 | — | 3,792 | gemini-3.8-flash-high; 95% CI [1303.54395703, 1324.61562115]; votes 3792; rank 8unknown | Original sourcewinner eligible | Third-party benchmark |
| #8 | Qwen3.8 MaxQwen | 1,313.60 | — | 10,017 | qwen3.8-max; 95% CI [1306.14335895, 1321.05468944]; votes 10017; rank 9unknown | Original sourcewinner eligible | Third-party benchmark |
| #9 | Gemini 3.5 FlashGoogle DeepMind | 1,309.62 | — | 13,103 | gemini-3.5-flash-high; 95% CI [1302.66440457, 1316.57227728]; votes 13103; rank 12unknown | Original sourcewinner eligible | Third-party benchmark |
| #10 | Muse Spark 1.3Meta | 1,309.35 | — | 4,550 | muse-spark-1.3-max; 95% CI [1299.58784189, 1319.11184130]; votes 4550; rank 13unknown | Original sourcewinner eligible | Third-party benchmark |
| #11 | Muse Spark 1.2Meta | 1,304.55 | — | 1,954 | muse-spark-1.2 (xHigh); 95% CI [1290.22866830, 1318.87305829]; votes 1954; rank 17unknown | Original sourcewinner eligible | Third-party benchmark |
| #12 | GPT-5.4OpenAI | 1,303.15 | — | 29,881 | gpt-5.4-high; 95% CI [1297.39164127, 1308.90995855]; votes 29881; rank 18unknown | Original sourcewinner eligible | Third-party benchmark |
| #13 | Gemini 3.6 FlashGoogle DeepMind | 1,297.66 | — | 8,127 | gemini-3.6-flash-high; 95% CI [1289.51471576, 1305.80526574]; votes 8127; rank 20unknown | Original sourcewinner eligible | Third-party benchmark |
| #14 | GPT-5.5OpenAI | 1,296.63 | — | 27,993 | gpt-5.5; 95% CI [1290.81697444, 1302.45204648]; votes 27993; rank 21unknown | Original sourcewinner eligible | Third-party benchmark |
| #15 | Gemini 3.1 ProGoogle DeepMind | 1,296.20 | — | 46,320 | gemini-3.1-pro-preview; 95% CI [1291.04958388, 1301.35691905]; votes 46320; rank 22unknown | Original sourcewinner eligible | Third-party benchmark |
| #16 | GLM 5.3 FlashZ.ai | 1,296.03 | — | 11,157 | glm-5.3-flash; 95% CI [1288.97978191, 1303.07121064]; votes 11157; rank 23unknown | Original sourcewinner eligible | Third-party benchmark |
| #17 | Claude Opus 4.8Anthropic | 1,294.18 | — | 18,341 | claude-opus-4-8-high; 95% CI [1287.71437228, 1300.65034925]; votes 18341; rank 24unknown | Original sourcewinner eligible | Third-party benchmark |
| #18 | Muse Spark 1.1Meta | 1,292.70 | — | 11,994 | muse-spark-1.1; 95% CI [1285.66303358, 1299.73172684]; votes 11994; rank 25unknown | Original sourcewinner eligible | Third-party benchmark |
| #19 | GPT-6.1 SolOpenAI | 1,288.19 | — | 1,813 | gpt-6.1-sol-max; 95% CI [1273.83724449, 1302.54208160]; votes 1813; rank 30unknown | Original sourcewinner eligible | Third-party benchmark |
| #20 | Grok 4.5xAI | 1,288.17 | — | 11,764 | grok-4.5; 95% CI [1280.98331591, 1295.35286239]; votes 11764; rank 31unknown | Original sourcewinner eligible | Third-party benchmark |
| #21 | Gemini 3 FlashGoogle DeepMind | 1,285.15 | — | 39,145 | gemini-3-flash; 95% CI [1279.92496783, 1290.36579718]; votes 39145; rank 32unknown | Original sourcewinner eligible | Third-party benchmark |
| #22 | Kimi K2.6Moonshot AI | 1,282.78 | — | 16,379 | kimi-k2.6; 95% CI [1275.98531996, 1289.57750584]; votes 16379; rank 33unknown | Original sourcewinner eligible | Third-party benchmark |
| #23 | Claude Sonnet 4.6Anthropic | 1,282.47 | — | 27,422 | claude-sonnet-4-6; 95% CI [1276.33534417, 1288.60341882]; votes 27422; rank 34unknown | Original sourcewinner eligible | Third-party benchmark |
| #24 | GPT-5.6 SolOpenAI | 1,281.39 | — | 10,907 | gpt-5.6-sol-xhigh; 95% CI [1273.98131322, 1288.80759068]; votes 10907; rank 35unknown | Original sourcewinner eligible | Third-party benchmark |
| #25 | GPT-6 AstraOpenAI | 1,281.35 | — | 3,590 | gpt-6-astra-max; 95% CI [1270.63592513, 1292.07239183]; votes 3590; rank 36unknown | Original sourcewinner eligible | Third-party benchmark |
| #26 | Qwen3.7 PlusQwen | 1,279.07 | — | 13,636 | qwen3.7-plus; 95% CI [1272.29971328, 1285.84102976]; votes 13636; rank 37unknown | Original sourcewinner eligible | Third-party benchmark |
| #27 | Gemma 4 31BGoogle DeepMind | 1,276.80 | — | 39,669 | gemma-4-31b; 95% CI [1271.07071523, 1282.52723839]; votes 39669; rank 38unknown | Original sourcewinner eligible | Third-party benchmark |
| #28 | Claude Sonnet 5Anthropic | 1,273.92 | — | 13,503 | claude-sonnet-5-high; 95% CI [1267.02186706, 1280.81063637]; votes 13503; rank 41unknown | Original sourcewinner eligible | Third-party benchmark |
| #29 | Qwen3.8 27BQwen | 1,271.39 | — | 7,013 | qwen3.8-27b; 95% CI [1262.99100972, 1279.79432635]; votes 7013; rank 42unknown | Original sourcewinner eligible | Third-party benchmark |
| #30 | GPT-5.6 TerraOpenAI | 1,271.07 | — | 11,020 | gpt-5.6-terra-xhigh; 95% CI [1263.65540928, 1278.48964371]; votes 11020; rank 43unknown | Original sourcewinner eligible | Third-party benchmark |
| #31 | Kimi K2.5Moonshot AI | 1,268.77 | — | 28,297 | kimi-k2.5-thinking; 95% CI [1263.09963340, 1274.44784290]; votes 28297; rank 44unknown | Original sourcewinner eligible | Third-party benchmark |
| #32 | Gemini 3.5 Flash LiteGoogle DeepMind | 1,268.15 | — | 8,381 | gemini-3.5-flash-lite; 95% CI [1260.07450401, 1276.22711331]; votes 8381; rank 45unknown | Original sourcewinner eligible | Third-party benchmark |
| #33 | Gemini 2.5 ProGoogle DeepMind | 1,263.29 | — | 88,805 | gemini-2.5-pro; 95% CI [1258.74589708, 1267.83852907]; votes 88805; rank 51unknown | Original sourcewinner eligible | Third-party benchmark |
| #34 | Grok 4.6xAI | 1,263.14 | — | 6,495 | grok-4.6-high; 95% CI [1254.48523007, 1271.79468461]; votes 6495; rank 52unknown | Original sourcewinner eligible | Third-party benchmark |
| #35 | Qwen3.5 397B A17BQwen | 1,262.52 | — | 32,058 | qwen3.5-397b-a17b; 95% CI [1257.19517602, 1267.83577721]; votes 32058; rank 54unknown | Original sourcewinner eligible | Third-party benchmark |
| #36 | Gemma 4 26B A4BGoogle DeepMind | 1,259.70 | — | 23,943 | gemma-4-26b-a4b; 95% CI [1253.23971386, 1266.15382555]; votes 23943; rank 55unknown | Original sourcewinner eligible | Third-party benchmark |
| #37 | Grok 4.20 Multi AgentxAI | 1,258.97 | — | 24,543 | grok-4.20-multi-agent-beta-0309; 95% CI [1252.84769478, 1265.09117861]; votes 24543; rank 57unknown | Original sourcewinner eligible | Third-party benchmark |
| #38 | GPT-5.6 LunaOpenAI | 1,258.24 | — | 11,086 | gpt-5.6-luna-xhigh; 95% CI [1250.82216689, 1265.66350676]; votes 11086; rank 58unknown | Original sourcewinner eligible | Third-party benchmark |
| #39 | MiniMax M3MiniMax | 1,253.11 | — | 19,026 | minimax-m3; 95% CI [1246.68252453, 1259.53765792]; votes 19026; rank 60unknown | Original sourcewinner eligible | Third-party benchmark |
| #40 | GPT-5.1OpenAI | 1,249.84 | — | 9,861 | gpt-5.1-high; 95% CI [1242.04728848, 1257.64065919]; votes 9861; rank 63unknown | Original sourcewinner eligible | Third-party benchmark |
| #41 | GPT-5.2OpenAI | 1,245.81 | — | 19,041 | gpt-5.2-high; 95% CI [1239.74622137, 1251.87027714]; votes 19041; rank 67unknown | Original sourcewinner eligible | Third-party benchmark |
| #42 | GPT-5.4 MiniOpenAI | 1,245.01 | — | 25,733 | gpt-5.4-mini-high; 95% CI [1238.81649273, 1251.21275945]; votes 25733; rank 68unknown | Original sourcewinner eligible | Third-party benchmark |
| #43 | Gemini 2.5 FlashGoogle DeepMind | 1,235.45 | — | 60,112 | gemini-2.5-flash; 95% CI [1230.89884927, 1239.99599852]; votes 60112; rank 74unknown | Original sourcewinner eligible | Third-party benchmark |
| #44 | Inkling SmallThinking Machines Lab | 1,235.32 | — | 7,472 | Inkling Small; 95% CI [1226.97211879, 1243.66134454]; votes 7472; rank 75unknown | Original sourcewinner eligible | Third-party benchmark |
| #45 | Grok 4.3xAI | 1,229.24 | — | 27,037 | grok-4.3; 95% CI [1223.34663531, 1235.13960477]; votes 27037; rank 78unknown | Original sourcewinner eligible | Third-party benchmark |
| #46 | Mistral Medium 3.5 128BMistral AISelected model | 1,223.08 | — | 5,375 | mistral-medium-3.5; 95% CI [1213.57754699, 1232.58266173]; votes 5375; rank 82unknown | Original sourcewinner eligible | Third-party benchmark |
| #47 | Mistral Large 3Mistral AI | 1,220.76 | — | 8,643 | mistral-large-3; 95% CI [1212.81813492, 1228.69335345]; votes 8643; rank 83unknown | Original sourcewinner eligible | Third-party benchmark |
| #48 | GPT-4.1OpenAI | 1,210.62 | — | 41,017 | gpt-4.1-2025-04-14; 95% CI [1203.95207689, 1217.29091390]; votes 41017; rank 87unknown | Original sourcewinner eligible | Third-party benchmark |
| #49 | GPT-5OpenAI | 1,208.24 | — | 34,470 | gpt-5-high; 95% CI [1201.43993030, 1215.03686763]; votes 34470; rank 89unknown | Original sourcewinner eligible | Third-party benchmark |
| #50 | GPT-5 MiniOpenAI | 1,202.14 | — | 28,959 | gpt-5-mini-high; 95% CI [1194.33907236, 1209.93452091]; votes 28959; rank 91unknown | Original sourcewinner eligible | Third-party benchmark |
| #51 | GPT-5.4 NanoOpenAI | 1,195.96 | — | 25,650 | gpt-5.4-nano-high; 95% CI [1189.64874058, 1202.27965668]; votes 25650; rank 94unknown | Original sourcewinner eligible | Third-party benchmark |
| #52 | GPT-4.1 MiniOpenAI | 1,181.35 | — | 40,221 | gpt-4.1-mini-2025-04-14; 95% CI [1173.87762706, 1188.81616531]; votes 40221; rank 101unknown | Original sourcewinner eligible | Third-party benchmark |
| #53 | GPT-5 NanoOpenAI | 1,159.22 | — | 3,929 | gpt-5-nano-high; 95% CI [1148.16831669, 1170.26403282]; votes 3929; rank 113unknown | Original sourcewinner eligible | Third-party benchmark |
| #54 | Llama 4 Maverick 17B 128E InstructMeta | 1,141.63 | — | 6,922 | llama-4-maverick-17b-128e-instruct; 95% CI [1132.58229133, 1150.67926500]; votes 6922; rank 120unknown | Original sourcewinner eligible | Third-party benchmark |
| #55 | Llama 4 Scout 17B 16E InstructMeta | 1,118.23 | — | 6,451 | llama-4-scout-17b-16e-instruct; 95% CI [1108.78959993, 1127.66568063]; votes 6451; rank 127unknown | Original sourcewinner eligible | Third-party benchmark |
| #56 | GPT-4o MiniOpenAI | 1,065.60 | — | 17,341 | gpt-4o-mini-2024-07-18; 95% CI [1057.29887437, 1073.89816601]; votes 17341; rank 135unknown | Original sourcewinner eligible | Third-party benchmark |
| #57 | GPT-4oOpenAI | 1,064.52 | — | 3,376 | gpt-4o-2024-08-06; 95% CI [1052.04133780, 1077.00060680]; votes 3376; rank 136unknown | Original sourcewinner eligible | Third-party benchmark |
| #58 | GPT-4.1 NanoOpenAI | 1,063.24 | — | 1,211 | gpt-4.1-nano-2025-04-14; 95% CI [1044.95158579, 1081.52393740]; votes 1211; rank 137unknown | Original sourcewinner eligible | Third-party benchmark |
58 models ranked by the best winner-eligible current-version score when available, otherwise the best publisher-artifact score; highest first. Observed Oct 19, 2026. The model you came from is highlighted.
Methodology and Coverage
Ranking protocolarena rating · rating. Each model appears once; verified winner-eligible runs take precedence, and missing models are not imputed.
Acquisition coverage58 published models from 165 source rows; 98 source identities remain quarantined.
Primary Evidence
Publisher Artifacts
Questions
LMArena Vision Arena FAQs
What does LMArena Vision Arena measure?+
Blind human preferences for multimodal responses in LMArena's vision arena. Model Markets classifies it as a multimodal benchmark and preserves the publisher's vision-2026-10-09-a64ca3b12b6e release as a distinct comparison cohort.
How are models ranked on LMArena Vision Arena?+
Models are ordered by arena rating in rating, with higher scores ranked first. Each model appears once; a verified winner-eligible current-version result takes precedence over a publisher-artifact-only result, and missing scores are not estimated.
Which model currently leads LMArena Vision Arena?+
Claude Fable 5 leads the current verified table with 1,323.95 rating on vision-2026-10-09-a64ca3b12b6e. This is a benchmark-specific result, not a universal model-quality claim.
How many models have a published LMArena Vision Arena score?+
58 catalog models are published from 165 source rows. 98 source identities remain quarantined rather than guessed.
Can LMArena Vision Arena scores be compared with other benchmarks?+
Raw scores should be compared only within the same benchmark version, metric, and protocol. The Intelligence ranking normalizes its three pinned publisher inputs to a frozen reference range, requires at least one source, and labels one- and two-source rows provisional instead of estimating missing inputs. View aggregate rankings →
Why might a model be missing from LMArena Vision Arena?+
A model remains absent when the publisher has no current result, the source model identity is unresolved, the evaluation protocol is incompatible, or the evidence cannot be verified. Model Markets does not substitute a provider claim or infer a score from a related model.
How current is the LMArena Vision Arena leaderboard?+
The current Model Markets snapshot was observed Oct 19, 2026 from LMArena artifacts. The exact publisher source and each retained result artifact are linked on this page.