Vision and multimodal

Results are only compared within a matching group, so a model tested one way is never ranked against a model tested another way.

Reason over images, audio, or video

Ranking groups use separate scales. Compare configurations only inside the same group; EvalRank never flattens unlike systems into one leaderboard.

ベンチマーク横断のコンセンサス

このトピックで採用されているベンチマーク全体でモデルを順位付けしたものです。意思決定のときと同じ重み付けを用いています。各スコアは、モデルの結果が最高結果に占める割合です。

RankModelShare of bestBenchmarks
1qwen38max100.0%1
2Claude Opus 4.799.9%1
3musespark99.5%1
4Claude Opus 4.699.5%1
5gemini3pro99.2%1
6GPT-5.598.9%1
7Gemini 3.5 Flash98.6%1
8musespark1198.5%1
9gemini31propreview98.4%1
10grok4598.4%1
11Claude Opus 4.898.4%1
12GPT-5.498.4%1
13GPT-5.2 Chat98.3%1
14gpt55instant98.2%1
15claudesonnet4698.1%1
16Gemini 3 Flash97.9%1
17glm53flash97.8%1
18gemini35flashlite97.5%1
19kimik2697.3%1
20qwen37plus97.2%1
21gemma431b97.1%1
22dolaseed20pro96.7%1
23Grok 4.20 Beta96.6%1
24Kimi K2.596.3%1
25mistrallarge496.3%1
26Grok 4.20 Multi-Agent Beta96.2%1
27gemini25pro96.0%1
28step5preview95.9%1
29qwen35397ba17b95.8%1
30mimov26pro95.7%1
31mimov26flash95.7%1
32qwen3827b95.6%1
33grok4395.6%1
34gemma426ba4b95.5%1
35GPT-5.195.4%1
36ChatGPT-4o Latest95.3%1
37kimik25instant95.1%1
38gemini31flashlitepreview95.0%1
39mimov2594.9%1
40minimaxm394.9%1
41glm5vturbo94.6%1
42gemini25flashpreview09202594.6%1
43GPT-5.294.5%1
44qwen35122ba10b94.3%1
45gpt5chat94.3%1
46GPT-4.5 Preview94.2%1
47ERNIE-5.0-Preview93.8%1
48qwen3527b93.6%1
49mimov2omni93.5%1
50o393.5%1
51gemini25flash93.5%1
52qwen3vl235ba22binstruct93.5%1
53GPT-4.193.4%1
54inklingsmall93.0%1
55claudeopus4thinking16k92.9%1
56claudesonnet4thinking32k92.8%1
57nvidianemotron35supervl120ba12bnvfp492.6%1
58o4-mini92.4%1
59GPT-4.1 mini92.4%1
60mistralmedium3592.3%1
61mistrallarge392.2%1
62Grok 4.1 Fast92.0%1
63claude37sonnetthinking32k91.7%1
64o191.7%1
65Claude Opus 491.5%1
66qwen3vl235ba22bthinking91.4%1
67Gemini 2.5 Flash-Lite Preview91.4%1
68Claude Sonnet 491.4%1
69Grok 491.1%1
70Claude 3.7 Sonnet90.4%1
71Gemini 2.0 Flash90.2%1
72glm46v89.3%1
73gemma327bit89.2%1
74Mistral Medium 3.189.2%1
75step1oturbo20250688.9%1
76Mistral Medium 388.9%1
77hunyuanvision15thinking88.8%1
78glm45v88.7%1
79hunyuanlargevision88.4%1
80llama4maverick17b128einstruct88.1%1
81step388.1%1
82Mistral Small 3.287.7%1
83Gemini 2.0 Flash-Lite Preview87.4%1
84llama4scout17b16einstruct86.7%1
85Claude 3.5 Haiku86.7%1
86Mistral Small 3.1 24B Instruct86.7%1
87step1ovision32khighres86.5%1
88qwen25vl72binstruct86.2%1
89qwen25vl32binstruct86.1%1
90molmo28b85.5%1
91GPT-4 Turbo85.5%1
92GPT-4o mini84.4%1
93Pixtral Large84.3%1
94GPT-4.1 nano83.7%1
95qwen2vl72b83.4%1
96Gemini 1.5 Flash-8B82.4%1
97step1v32k81.9%1
98Claude 3 Opus81.7%1
99Molmo 72B80.4%1
100Hunyuan Standard Vision80.1%1
101llama3290bvisioninstruct79.4%1
102qwen2vl7binstruct79.3%1
103Pixtral 12B78.9%1
104internvl226b78.8%1
105amazonnovalite1078.4%1
106amazonnovapro1078.3%1
107Claude 3 Sonnet78.2%1
108yivision77.1%1
109ayavision32b77.0%1
110Claude 3 Haiku76.9%1
111Molmo 7B-D76.4%1
112llama3211bvisioninstruct76.4%1
113nvilainternal15bv175.8%1
114llavaonevisionqwen272bov75.2%1
115llava1634b74.2%1
116cogvlm2llama3chat19b74.1%1
117minicpmv2674.0%1
118internvl24b73.7%1
119phi35visioninstruct70.7%1
120phi3vision128kinstruct67.7%1

アリーナシステム · 群衆によるペアワイズ比較

Exploratory - not a ranked winner yet
rg-vision-multimodal-arena-system-crowd-pairwise-arena-system-v1

エクスプローラーのエビデンス

LMArena Vision

単一の出典。に確認しました。まで有効です。

RankConfigurationCapabilityUncertaintyFamilies
1Claude Fable 5 (High) 2026-10-19config_2fc50238a92d9751ea35074c52b2170d51985f2f6b9554d749ac80b8ef040f09100%不明1
2Qwen3.8-Max 2026-10-19config_b105af9a815223ae9705a1d89539f8c84f54766ab728185de10094212176847d98%不明1
3Claude Opus 4.6 (High) 2026-10-19config_6d8411e4939d2152abe36e4b736f8be9faa4e72fd29f7fc9f084738dfac6553398%不明1
4Claude Opus 4.7 2026-10-19config_9f33bff2e9728c30da5a8a7d26e1489700f226254261768deab8e62116112a6598%不明1
5Claude Opus 4.7 (High) 2026-10-19config_48b171f2ffee9f19174ba19017c72151d566d6dbb734901e1b2759428cf59cc297%不明1
6Gemini 3.7 Flash (High) 2026-10-19config_3edb5fd33c7c59583d85a511f565d0818f0b079bc899b8ece6de7c67f5194ff697%不明1
7Muse Spark 2026-10-19config_fb01e2dc1a7d67463f16115a1bce067f147b8ec4f1d64258cfe7da8b45d70d5197%不明1
8Claude Opus 4.6 2026-10-19config_bab0c8c449fcaf56513a6602ddbc4af423aed9954a087048e233dead1416d49996%不明1
9GPT-6.1 Sol (Max) 2026-10-19config_7d85adf2b6a0d71e8b1fc8472a6433cc5eb1f82aefcd93cfa4e07aa69a72fa5b96%不明1
10Claude Opus 5.5 (High) 2026-10-19config_d2270b566111eda3ea91970277701681b96f942841bdda14eda26c09709ff13896%不明1
11Muse Spark 1.2 (xHigh) 2026-10-19config_902333faeb7bfc67086b5df509c7e02a17feb726474f27a620f589975da5446196%不明1
12Gemini 3.8 Flash (High) 2026-10-19config_b4dfa48e6a3a36778090a5d7a5bad975d490a4b9320434dc752070a1e62011fc96%不明1
13Muse Spark 1.3 (Max) 2026-10-19config_41ab01beb8f9ec4b5ee18cbf10a8273c08728c23e48b119a8fbb63519efda58d96%不明1
14Gemini 3 Pro 2026-10-19config_c61dd6afd86a0e9dfe10ce50c5e354073c2c08347c215198cd30bdf6836de33a95%不明1
15Claude Opus 5 (High) 2026-10-19config_21026b0ab42efd5568be8b3bfcf55556f1907384987dfa333ba50f2a3feb9db995%不明1
16Muse Spark 1.3 (xHigh) 2026-10-19config_665ce262c9d658d7a5a13186f9b3f975ed2b30b6d79639cfca0d4f1aedd2c4da95%不明1
17Claude Fable 5.1 (Max) 2026-10-19config_787a17c19fd1fdebe427718ce60522e2c9ce0a1493035b3bfd8e5462e10be69d95%不明1
18GPT-5.6 Sol (xHigh) 2026-10-19config_e88e8e1387d0bb4b04a6d1a86686a5e1fcc59a710e824d72708ae1780d34b3d795%不明1
19GPT-5.5 2026-10-19config_db86194517234f3675cbd45885960f81d2f7299de6b5a273ca30d8788458cf8d95%不明1
20Claude Opus 4.8 (High) 2026-10-19config_a4daf4a5a4a0906e5aaf77fee4800c38f21f550c091f264550f120af0e15652795%不明1
21GPT-6 Astra (Max) 2026-10-19config_645dd67b7efce10b3447c350253bf037653146af18744060f363ef81396a054394%不明1
22GPT-5.4 (High) 2026-10-19config_971e46ec45be374bd6d262aad6b20be3454470cc560c34f5489b8f543b43cf8994%不明1
23Gemini 3.5 Flash (Medium) 2026-10-19config_65136762e12683832c3065f7d8a13c361d6fd85737445522566960cc3ce5d24794%不明1
24Gemini 3.5 Flash (High) 2026-10-19config_6a933328b835027963256e1b6c598a57c1bdaca381aa6bc8dbb1ced5d0f250d494%不明1
25GPT-5.5 (High) 2026-10-19config_81e9ed2bac02c59d85f0fc7bc3a17533cc7974fd9063eb6820831dbd500e3d9094%不明1
26Gemini 3.6 Flash (High) 2026-10-19config_f02ed696019adc8cfb5b45fd657e2572e694066a6f9ea86a4fddc9a77565ca7594%不明1
27Muse Spark 1.1 2026-10-19config_dc45fa8470327dcec96432c732ecc82934b3a669ebd5be742534f63e87ac22a693%不明1
28Gemini 3.1 Pro Preview 2026-10-19config_a8e2c48295ea11c5414b6451e18d6591b49f339a2bd0e5bf33f4dede66e8987693%不明1
29Grok 4.5 2026-10-19config_bed5bf7b482e11ec61e30621c5426a70dbc59ea4e125ee40fa9c97547e197a0293%不明1
30Claude Opus 4.8 2026-10-19config_81c00d79961d026114d90a28589fdb0cf8ba67a5cf60156c2bef3137e141ff3c93%不明1
31GPT-5.4 2026-10-19config_d2f49de5efaa295f786e1731d1c71d6cda6137d548d658d28d343478115ca33093%不明1
32GPT-5.2 Chat (20260210) 2026-10-19config_2a9c259ca180eb5a2a713e0623add152921ebcf88fd14648d68e913cfec207d693%不明1
33GPT-5.5 Instant 2026-10-19config_518cdb922ba4dfce80b7c3d9b2574931e8994d814e3f163c2382f3bec18da83d92%不明1
34Claude Sonnet 4.6 2026-10-19config_21baa420b0c5e01da3d7d51cbd4c84586d0f90ff3747ee2b652619fc3d87bad292%不明1
35Gemini 3 Flash 2026-10-19config_95304c5e5f2986652c04ff9243fb34444e0b37e378572580893ca9737998044c92%不明1
36GLM-5.3-Flash 2026-10-19config_927a2f6f1f0daebce146fd217ac0fec29d2cd3064b421fc4fe161a932f35e2a991%不明1
37GPT-5.6 Terra (xHigh) 2026-10-19config_a9b30385589f6e5748d8d0977d8f4855bbab8befb3cf3ca5282d0348f094aa1e91%不明1
38Gemini 3.5 Flash-Lite 2026-10-19config_d102e21062bfb8938ded8259742b9bc00f1e209ee1c882739c723451d8e070ff90%不明1
39GPT-6 Sol (Max) 2026-10-19config_3b2c63f79cf0916c0c180cde0fc112cad5bc978b90ea5481bff4c1b3267b65fa90%不明1
40Claude Sonnet 5.5 (xHigh) 2026-10-19config_b1c8e0bc6fed9955c228bf0fc18e6e8460b3b22f25711dd5b3c96c76ae8bd33090%不明1
41Kimi K2.6 2026-10-19config_55c279146e618783eb8fe91b6f28bc5d1623ab3327e4affeeae577cb1c1c9c0990%不明1
42Qwen3.7-Plus 2026-10-19config_f0f0fb9039ef2a66ae7315d0316dadd17bb92a901ad3618ab7fcad25c57e4f0789%不明1
43Grok 4.6 (High) 2026-10-19config_869165bb1837c4958c6ced5fab3d45f0f0674d281157c098216477fc7f5b2b0989%不明1
44Claude Sonnet 5 (High) 2026-10-19config_7420059a038e5ce65d88a5dcfb8580a931738b46936f4018eb95ed671763bd7e89%不明1
45Gemma 4 31B 2026-10-19config_ac64e5545a93de1f466271e1f5a93c335ac4eb5318a3e3700f03ceadd14ec15b89%不明1
46DeepSeek-V4.1-Flash-20260910 (Max) 2026-10-19config_7631a77ef47c13fb9826a4f024119daff3dd37291217aaa77eb66b0b1da7d02e89%不明1
47Gemini 3 Flash (Minimal) 2026-10-19config_7b11f2ea36e1aae44e6f160fce16500df20de952d7a19b8d566da4dc893f7d3889%不明1
48GPT-5.6 Luna (xHigh) 2026-10-19config_d2eb9213ccfb77a43836c854622c63c1666cfac22c465e571e1aab66204da1ec88%不明1
49Dola Seed 2.0 Pro 2026-10-19config_af7c915b37ee636d7d01ec1407080ab9857f7240dcd6dc13e61a9a31152c98c388%不明1
50Grok 4.20 Beta (Reasoning, 0309) 2026-10-19config_4cbd254b6497b4a914a42299384f37cf951af098dd41d8e9faf2f8e73109819687%不明1
51Kimi K2.5 (Thinking) 2026-10-19config_456ccd430ee7d07a786f6e9074f251c36e47546caf4b11e989316a9ef942d4a787%不明1
52Mistral Large 4 2026-10-19config_cd198923343b49182379e92c73f2f3c448673c16d95cf7b9c043432c6ff294b887%不明1
53GPT-5.1 (High) 2026-10-19config_10e3c3c03a24cb0b81d662a5bebe9b5d43f8d255d3dd55ccaa7e9f5544b8bde487%不明1
54GPT-5.4 mini (High) 2026-10-19config_5423d05a894248acb259232a4cb87f524259ec54ca9c1031fe59ad5ac3c7d6b686%不明1
55Grok 4.20 Multi-Agent Beta (0309) 2026-10-19config_87ac146c6df6f797bfc5a8b13b0fd3322bbdadde54ee39213525c43cec19a12386%不明1
56Grok 4.7 (xHigh) 2026-10-19config_d1a8813a7c53391ef224874f9c75440bf5022025b59a71d2a2ab90546ff9cda586%不明1
57Gemini 2.5 Pro 2026-10-19config_7e75d2d5fe38a54fc822cb223458b25c9132437123ea18c7eed03c1089ce7d9b86%不明1
58Step 5 Preview 2026-10-19config_035907a374463d3a92513ba12ad61a5184cbf1d41c9b613fb8bb1abba907085d86%不明1
59Qwen3.5-397B-A17B 2026-10-19config_1ee3035bb0555062b8e321f19c9adb59db117772e316453977d547c951c3bcbd85%不明1
60MiMo-V2.6-Pro 2026-10-19config_ab88f747fd7815e574b5a88fe60c94647f7fc68bbeb58405364cc1f2c0b7137285%不明1
61MiMo-V2.6-Flash 2026-10-19config_6bb2ad59b064d0214a4d34d0675c466d40b58df21848334fd6fee16d66224c8a85%不明1
62GPT-5.2 (High) 2026-10-19config_fc4ca7ea2be1f4dd6fcf9e5f8dc50f41674a78371a1a3e81308dd75054b2e6c285%不明1
63Qwen3.8-27B 2026-10-19config_50485d9aaab49abd97ea8ff4ca0358d6604f96219557fc44570846271ecf45f185%不明1
64Grok 4.3 2026-10-19config_9c051d62c2f39ba3412fde0a2f4ebe98ceaac8d73b0c449105e422c072728caa84%不明1
65Gemma 4 26B A4B 2026-10-19config_6ed7fe999a443dee3b042adc547a8a8ac30a3c0d08ab387758176591fdbed7c184%不明1
66GPT-6 Luna (Max) 2026-10-19config_bf7516a58d6b6889c753e01bbccd8fd33b2e8a1ff4fd88bdea49d7a8706c6d6084%不明1
67GPT-5.1 2026-10-19config_69bf9059e5f2be8010ee92d44b6a6edae5b6ac001c3215b10853b5973fd14dd684%不明1
68ChatGPT-4o Latest (20250326) 2026-10-19config_c14e6967a70a1afd1342d82bc7a5fb9167e81e706d4f45b68e6214b45657b2ec84%不明1
69Kimi K2.5 (Instant) 2026-10-19config_bac6aecf2441e1b551bbdca25b6914a4818c52de6ffe842b7b700e48b5b57d4a83%不明1
70Gemini 3.1 Flash-Lite Preview 2026-10-19config_585fd60a331a42040ad44fad1d86c113af57b58dc54b129086c0843e8d7c9f5b83%不明1
71MiMo-V2.5 2026-10-19config_2e83fc64578de0f5661dadce8071f29bd756ca726de56b5f3abcde1fc60f9e1082%不明1
72MiniMax-M3 2026-10-19config_be143d629315e35425d6a882bb4e9889ab240dde10d1d9e1c419dcd532a181b382%不明1
73GLM-5V-Turbo 2026-10-19config_66c4812a93013c8473c1efb8b3386043e8d0e549c96f7cc2c5f115e92008f18982%不明1
74Gemini 2.5 Flash Preview (09-2025) 2026-10-19config_8b69f56fe727c2eebb1b3bad2c56f28eb9c12854f22525c2a0bef9ccc4f3caff82%不明1
75GPT-5.2 2026-10-19config_3d8cb1b5ee5937f08eeada2b6a174a49cf1db17470ee76fbab64c3e1b711cc8e81%不明1
76Qwen3.5-122B-A10B 2026-10-19config_a02d808c61eac76bece0d07266cfbaa6776f1743295799a880aa3c50c24bfd3181%不明1
77GPT-5 Chat 2026-10-19config_b73c099a7438eea274b1d4ce7d3a97e0f4c1c78694b27ffcba7a86378762812d81%不明1
78GPT-4.5 Preview (2025-02-27) 2026-10-19config_0d096f0eac7f0dcd9e6805bbd9f252162214be1ff0183a64658c7eb8e3ea008b80%不明1
79ERNIE-5.0-Preview (1220) 2026-10-19config_e2864698e8ccf4a01a7e59febf48f18afc3e3870b943d6310fc13c396f5a709c79%不明1
80Qwen3.5-27B 2026-10-19config_4ca61083c6b1ecafd0e1d690d0f65a1336abf75b63a732c241a593d45221a90c79%不明1
81MiMo-V2-Omni 2026-10-19config_68663a0b7f73eddaa6c71098205520404de4c2fdbac858ab061228436f0a888e78%不明1
82o3 (2025-04-16) 2026-10-19config_1cc9dcdaef569d3d2fb0fa95a4555fdfde88bbc9794701643189b29220f0cca878%不明1
83Gemini 2.5 Flash 2026-10-19config_10d3ea9d3a7c3d17b5b32864d016a4f17e443385931e23233db250d65587f12a78%不明1
84Qwen3-VL-235B-A22B-Instruct 2026-10-19config_ab8817c0360bdcdf834fe4f3a972615d4f65899595fe0f4124b1798a5762be8e78%不明1
85GPT-4.1 (2025-04-14) 2026-10-19config_7903b63773b091849150121223a7d4c947b16d2b435412e66239e8f970df181578%不明1
86GPT-5 (High) 2026-10-19config_da1b2bd3d6220a0396df561081f37d69f98b3647bedde0d14f62b884b3eff6ce77%不明1
87Inkling-Small 2026-10-19config_56eeed4b8a0e3a567009d3df584390ec38b42cc25e8c3c15881cdb6b62cc5b7677%不明1
88Claude Opus 4 (Thinking 16K, 20250514) 2026-10-19config_53f5ff7d128471f7f45e587af4a6ab97764ac640b7afb47ee3d47e11fadd061176%不明1
89Claude Sonnet 4 (Thinking 32K, 20250514) 2026-10-19config_6f1a4d6457ab02db2c53c0bbef54cabd4739d57a1afc67e89f59055f3542626676%不明1
90NVIDIA-Nemotron-3.5-Super-VL-120B-A12B (NVFP4) 2026-10-19config_5102bc54e266322973b6f78cadd208fe0acaa736c76d78be57ffc783ba52a1c876%不明1
91o4-mini (2025-04-16) 2026-10-19config_948b08edab93f06d7471a91af0a5208357d26d29ac541ee1c5f5257a51d001db75%不明1
92GPT-4.1 mini (2025-04-14) 2026-10-19config_ba285eda0e8dadc7da283efe083508f2071d660fe4236b051b3e0676330b182975%不明1
93Mistral Medium 3.5 2026-10-19config_d3f8ab82e750172f67f1c3d77b072762322ee6b596eb0973500c8c44f14eee8075%不明1
94Mistral Large 3 2026-10-19config_f9ad73d46fcc4c01373e6ab7256d5ba4086c0d37ab60bc94c24d04d04fb6127374%不明1
95GPT-5.4 nano (High) 2026-10-19config_0bdd8c29c34ca97755ae0d4ddee16f7ff94e81e3bb19b707cb9007d18dd399e574%不明1
96Grok 4.1 Fast (Reasoning) 2026-10-19config_40c0aefa96afe67818fe5f6db46a1ea92a10be0ff2d9f783f15fb4c95156ab9774%不明1
97Claude 3.7 Sonnet (Thinking 32K, 20250219) 2026-10-19config_d095f498f51067b5ddc63463172ed3a3174336375744bd663a572c742dc2b71573%不明1
98o1 (2024-12-17) 2026-10-19config_5020c09f8f3a7b9e3cd223c7fffecadbdb9b610969e1b3cd9ee94b00e9676e3c73%不明1
99Claude Opus 4 (20250514) 2026-10-19config_19e4279b5231dbf64740ee4a68cb821462648a0938be44e88469911b3377683272%不明1
100Qwen3-VL-235B-A22B-Thinking 2026-10-19config_a3a7ebe47c45870f84add22c1ea429de823a70c4f9dfeb1140d08b0142359e9272%不明1
101Gemini 2.5 Flash-Lite Preview (Thinking, 06-17) 2026-10-19config_1849471d43a96b42eee5d047d104550245861fe6105a2276115cd4e256df87a972%不明1
102Claude Sonnet 4 (20250514) 2026-10-19config_7037f17fb0c87089aef35e05613c5ec7117a6fd4685ed045471ac404104006a772%不明1
103Qwen-VL-Max (2025-08-13) 2026-10-19config_591db8f933a728876ec8efd5352391d174d479cd995ac8717e15122598ce30fb71%不明1
104GPT-5 mini (High) 2026-10-19config_0bf984f4ea3929cdd9542925f6c31520d3447df3477cf3e0c62f4b37c1a8a64a71%不明1
105Grok 4 (0709) 2026-10-19config_21ba5aa79d0c747b42129bde2eaf8dbb9b8c8100b57ea887ae817ee6b1aba66e71%不明1
106Gemini 1.5 Pro (002) 2026-10-19config_fa221cf716b5090aff6c640ac9f690c763f2f1f6cd76a7cdb92a07dd26fd280070%不明1
107Claude 3.7 Sonnet (20250219) 2026-10-19config_5ee21e1d63c1c9934d85f4b91bd3b030458afd9742b924fcc7174b68bd0dc6aa69%不明1
108Gemini 2.5 Flash-Lite Preview (Non-Thinking, 09-2025) 2026-10-19config_d8423f42ad3a301211dfe6d953b5b24cecac38dab0efbe5da2a656971a60c5ad69%不明1
109Gemini 2.0 Flash (001) 2026-10-19config_c524c9a346e46ff4e230f198addd65194351c1223d422067c89134a81fadac6068%不明1
110GPT-4o (2024-05-13) 2026-10-19config_ff5a31ead743dd5a338fc2f134f000f4d7ad318a391aeecc20171a452ce8551666%不明1
111GLM-4.6V 2026-10-19config_410cf05bc3a3be17f41ba20ec01f0d31e379a77054e2c936d985983ee0d0542d66%不明1
112Claude 3.5 Sonnet (20241022) 2026-10-19config_fe2a94627b4b39826708a5a4f70812c45754e4d5391d8a14eb8afe1997587d0565%不明1
113Gemma 3 27B IT 2026-10-19config_24bf14df3559c005a9454d83073d8deb89f28673cd3a9873266d688f437be23765%不明1
114Mistral Medium 3.1 (2508) 2026-10-19config_fd2eddb4cc95fa25b6700bda6aa6a825f6f1a92a15d105a0a9a07e8fe408e6ab65%不明1
115Step-1o Turbo (202506) 2026-10-19config_164772853d8e29028ab11102af314dd2509ffc8b8cc78013f203a49c3348a29c64%不明1
116Mistral Medium 3 (2505) 2026-10-19config_d959a5227b95a103ff44822a12938d6f09ec68e803e149993f75c7fa77f5572764%不明1
117Hunyuan Vision 1.5 Thinking 2026-10-19config_57dc93aed8e256cba7173f57ce6c7c1a50d354105fa8767169c97e19209e0bb864%不明1
118GLM-4.5V 2026-10-19config_ea76023630ee83fcef731dcd527c445f8ce484e1d104dbc343a2c611e32015f964%不明1
119Hunyuan Large Vision 2026-10-19config_8585606707681fdb235b990f8d7fd192185ec5e6c9517c556b68cda297a0125463%不明1
120GPT-5 nano (High) 2026-10-19config_306d82d4532c2f817e9145025136ce85d7b769b1f468d9c5a8ba74ef53facf0962%不明1
121Llama 4 Maverick 17B 128E Instruct 2026-10-19config_7eac998082a22bd41afba4277396eb3e94977df68ade374adbcb3292e82dcbb262%不明1
122Claude 3.5 Sonnet (20240620) 2026-10-19config_f545d7e456d69f29f20648787be122805694c3b1b7746616b376fc5dbd9dae7462%不明1
123Step-3 2026-10-19config_c53e896e93332ae6c85a738e91382ed375f868e357e980fe97169d8017a6637f62%不明1
124Gemini 1.5 Flash (002) 2026-10-19config_4a93b12bb13f5e4b06e47b259c815894c6baca314d73046c8d066d84408fdd1661%不明1
125Mistral Small 3.2 (2506) 2026-10-19config_678fd00a66c940dbe7121251a533f7e3686c7bf77fe6bca9fa634cbc9de882d261%不明1
126Gemini 2.0 Flash-Lite Preview (02-05) 2026-10-19config_6350fce6cc2617800f99ad7e7e09576898ad47bd8f894f610772242a9c3b058e60%不明1
127Llama 4 Scout 17B 16E Instruct 2026-10-19config_3d8d9cc44fc57ea5b9369029917234e092d9e5409f441716e323d5404cdd5e3c58%不明1
128Claude 3.5 Haiku (20241022) 2026-10-19config_438e904df3fd2edfecc22253e5ad01312416ae7b8d3c7c43d9475e541794eee858%不明1
129Mistral Small 3.1 24B Instruct (2503) 2026-10-19config_53bc59d34727eb210dfcab711f25b88b270731604ce53074bd58124d6d398d1d58%不明1
130Step-1o Vision 32K (High Res) 2026-10-19config_1ef603aef13beac12159bbebae4d51aa0415d30c96d6472c314aed57dc2e3a3e57%不明1
131Qwen2.5-VL-72B-Instruct 2026-10-19config_14f59602bfe535fd6ab754c2d942d8bf2032b72702a8f2452b9d519912e1c1bb56%不明1
132Gemini 1.5 Pro (001) 2026-10-19config_a8b2c6e8d0ece3211a013701058c21e54f8380afb79c8695df7994ae8630e11456%不明1
133Qwen2.5-VL-32B-Instruct 2026-10-19config_59433fb4ede9e0ae2f82b8af70a00e5277c6f61742cbcde8e2e877d25ebea1a256%不明1
134GPT-4o (2024-08-06) 2026-10-19config_bfc4018d875d438cd52d6751f5c62c6f7f5537db666320faea04e8e1b82928f755%不明1
135Molmo 2 8B 2026-10-19config_039b1ec85b314470d571dc99b587fb4f9387b083840dbb0fb3c9efb7f70b295654%不明1
136GPT-4 Turbo (2024-04-09) 2026-10-19config_3f1542bd66650abafb9cdd00f33610b53b054f788f7d4afa9f76c612e9cc83b854%不明1
137GPT-4o mini (2024-07-18) 2026-10-19config_8e134bef46d1a23539b3ab5270bf91173e835a16ecfca3102208204545468c5951%不明1
138Pixtral Large (2411) 2026-10-19config_4b32e7072a935093f6384b4dd79bdaaa4bd0c95169158ee163385886fd842eb950%不明1
139GPT-4.1 nano (2025-04-14) 2026-10-19config_c410aea9e08a017c24ef1d21afbc31e195d23250d0a12cb9e5c0016390d11b3e49%不明1
140Qwen2-VL-72B 2026-10-19config_c3105032baab2dad5457c6c59866495dde4b3046a8e7eb91620e7f9295a1523248%不明1
141Qwen-VL-Max (1119) 2026-10-19config_12f1584d06ea668241859d32b1002cabd14e6e6cf4b86b4b542b077d10964d9348%不明1
142Gemini 1.5 Flash-8B (001) 2026-10-19config_aecacfb23e707bee1162c1da1804cdc3f56fff5a5ca537d5312549e8a442184445%不明1
143Step-1V 32K 2026-10-19config_5eb6a5da56db9c964604a4bd88a885002bcf9549f9cc2a6197692beada22299043%不明1
144Claude 3 Opus (20240229) 2026-10-19config_7c5848f0617c09ba1f1321877530f6561cab3b1686ecbc245a331ce13266716342%不明1
145Gemini 1.5 Flash (001) 2026-10-19config_1f7e4ff8a064eeb8423834ae32c2f0d34bc198882499e8867b05b7d17f1cca2742%不明1
146Molmo 72B (0924) 2026-10-19config_0a649244127051cd020cae5b7912027502eee98aaca520c1ba7e56325dd52a6938%不明1
147Hunyuan Standard Vision (2024-12-31) 2026-10-19config_35ce27f521ff36c22001c1937a8e58f5d4c6c3e51ab24cfd7833d47fb0d1740438%不明1
148Llama 3.2 90B Vision Instruct 2026-10-19config_bd7069238a3f0278ee87c52476a4025c4903761bc34120e1a25a9723a17d14da35%不明1
149Qwen2-VL-7B-Instruct 2026-10-19config_3e7207188d0f0f32789b1bd30a16289a455fced3ee29e17e5e9769ff112e8c3635%不明1
150Pixtral 12B (2409) 2026-10-19config_05cfc09a44411d7b1674da3348ef8e62f36593b81004d531a8fea25d5504823c34%不明1
151InternVL2 26B 2026-10-19config_03129bef035e1e8c1a315339d76de0383edf382a6c9fc2b5b332afb3f2e012a534%不明1
152Amazon Nova Lite 1.0 2026-10-19config_a462d4006bd9cdd1c740d8320564cffcc1ab7a465b911f10f47e8f5065e1db7132%不明1
153Amazon Nova Pro 1.0 2026-10-19config_4ae586bfde51333130f6957900efde9b4917c55990dd4e386623eb826ee552e332%不明1
154Claude 3 Sonnet (20240229) 2026-10-19config_981f29746b86668ad3ba7bd09f9a29703379da0c59329d24dba421300235d65732%不明1
155Yi Vision 2026-10-19config_d659a6aa2a14be08f69a824b94e4f32b3ba3035fe184b6d249370253ca7fe8e128%不明1
156Aya Vision 32B 2026-10-19config_85328779a80a8d27099872be9f72dc5bd672289f4c71383ba3794526f616a9ee28%不明1
157Claude 3 Haiku (20240307) 2026-10-19config_abd11ce67e5b71c6f0ec18a3f6e8907e3cdf13b63a94c46f86e707a48d6890c628%不明1
158Molmo 7B-D (0924) 2026-10-19config_b97d2d14c0f931d4014523ec211dff346e085bd39d8682ceee53ba216222922126%不明1
159Llama 3.2 11B Vision Instruct 2026-10-19config_941b0e5bbd9888959cfbd5b99917a71f11e7830646bd939f76160c596a7b248126%不明1
160nvila-internal-15b-v1 2026-10-19config_f05a336a38b52cb53bc64fb60530e0a748b34d784d45a5192e12c3a7e0a6d3d824%不明1
161LLaVA-OneVision Qwen2 72B OV 2026-10-19config_9819da4c1be8189b56bab9a71ca5470ef42563260e998f07562f5c5789d6eb1023%不明1
162LLaVA 1.6 34B 2026-10-19config_ba018229fd45b2614eeba971b467a964ae7c1039567caab4e3d71678a1c8903220%不明1
163CogVLM2 Llama3 Chat 19B 2026-10-19config_e4d2f1071637fb720bbe863579fc216d36147f17bcae458cbf5c04b2631817b819%不明1
164MiniCPM-V 2.6 2026-10-19config_b5b28f800df2a28f4e3205dbee3e762f87c5785d38eace9a628a34cc201c0ed919%不明1
165InternVL2 4B 2026-10-19config_6da16b8bf45cbb22633960b0ecfd2c617bf2e2101ed729515cd395182958c14b18%不明1
166Phi-3.5 Vision Instruct 2026-10-19config_3507694f567f91e165e1b6c836f60a40b2ba1652650d9d31d311bcfc27065b479%不明1
167Phi-3 Vision 128K Instruct 2026-10-19config_2156dca419dec8ad5117075173f3dbc8ebc4fed64fb3d9f0901f7241b58211130%不明1
このグループの構成を比較する
上位セットを定めるための独立したエビデンスが不足しています。
  • キャリブレーションが検証されていません。
  • 独立したベンチマークファミリーのカバレッジが基準値を下回っています。
Back to all kinds of work