Image generation
I risultati vengono confrontati solo all'interno dello stesso gruppo, così un modello testato in un modo non viene mai messo in classifica contro un modello testato in un altro.
Generate images from a text prompt
I gruppi di classifica usano scale diverse. Confronta le configurazioni solo all'interno dello stesso gruppo; EvalRank non appiattisce mai sistemi diversi in un'unica classifica.
Consenso tra i benchmark
Una classifica dei modelli sui benchmark ammessi per questo argomento, con lo stesso peso che si dà alle decisioni. Ogni punteggio è la quota del miglior risultato ottenuta dal modello.
| Posizione | Modello | Quota del migliore | Benchmark |
|---|---|---|---|
| 1 | gptimage25sunburst | 100.0% | 1 |
| 2 | gptimage25flare | 98.2% | 1 |
| 3 | GPT-Image-2 | 96.8% | 1 |
| 4 | maiimage26 | 93.1% | 1 |
| 5 | gemininanobanana21 | 92.8% | 1 |
| 6 | reve21 | 91.1% | 1 |
| 7 | museimage | 88.9% | 1 |
| 8 | reve20 | 88.8% | 1 |
| 9 | gemini31flashimagenanobanana2websearch | 88.1% | 1 |
| 10 | qwenimage30pro | 87.9% | 1 |
| 11 | maiimage25 | 87.8% | 1 |
| 12 | seedream50pro | 87.6% | 1 |
| 13 | gemini31flashliteimagenanobanana2lite | 87.5% | 1 |
| 14 | gemini3proimagenanobananapro2k | 87.3% | 1 |
| 15 | gptimage15highfidelity | 86.6% | 1 |
| 16 | gemini3proimagepreviewnanobananapro | 86.3% | 1 |
| 17 | qwenimage21 | 85.8% | 1 |
| 18 | ideogram40quality | 84.1% | 1 |
| 19 | Qwen-Image-2.0-Pro | 83.5% | 1 |
| 20 | uni11max | 83.3% | 1 |
| 21 | maiimage2 | 82.9% | 1 |
| 22 | uni11 | 82.8% | 1 |
| 23 | grokimagineimage | 81.9% | 1 |
| 24 | recraftv41utilitypro | 81.8% | 1 |
| 25 | cosmos3supertext2imageagentic | 81.6% | 1 |
| 26 | grokimagineimagepro | 81.4% | 1 |
| 27 | flux2flex | 81.0% | 1 |
| 28 | revev15 | 80.9% | 1 |
| 29 | flux2pro | 80.8% | 1 |
| 30 | hunyuanimage30 | 80.6% | 1 |
| 31 | Cosmos3-Super-Text2Image | 80.6% | 1 |
| 32 | seedream45 | 80.5% | 1 |
| 33 | gemini25flashimagepreviewnanobanana | 80.5% | 1 |
| 34 | imagen4ultragenerate001 | 80.3% | 1 |
| 35 | flux2dev | 80.2% | 1 |
| 36 | seedream50lite | 79.8% | 1 |
| 37 | seedream402k | 79.8% | 1 |
| 38 | wan26 | 79.7% | 1 |
| 39 | recraftv41pro | 79.3% | 1 |
| 40 | imagen4generate001 | 79.0% | 1 |
| 41 | krea2medium | 79.0% | 1 |
| 42 | wan25preview | 78.2% | 1 |
| 43 | seedream40fal | 78.2% | 1 |
| 44 | krea2large | 78.1% | 1 |
| 45 | gptimage1 | 78.1% | 1 |
| 46 | recraftv4 | 78.1% | 1 |
| 47 | hidreamo1image | 78.1% | 1 |
| 48 | krea2turbo | 78.0% | 1 |
| 49 | seedream40highresfal | 78.0% | 1 |
| 50 | gptimage1mini | 77.7% | 1 |
| 51 | wan27imagepro | 77.3% | 1 |
| 52 | wan27image | 77.1% | 1 |
| 53 | recraftv41flash | 76.9% | 1 |
| 54 | maiimage1 | 76.5% | 1 |
| 55 | zimageturbo | 75.8% | 1 |
| 56 | seedream30 | 75.8% | 1 |
| 57 | flux2klein9b | 74.9% | 1 |
| 58 | qwenimagepromptextend | 74.3% | 1 |
| 59 | flux1kontextpro | 74.2% | 1 |
| 60 | imagen3generate002 | 74.1% | 1 |
| 61 | ideogram30quality | 73.5% | 1 |
| 62 | lumaphoton | 72.6% | 1 |
| 63 | pimage | 72.3% | 1 |
| 64 | flux2klein4b | 72.1% | 1 |
| 65 | runwaygen4 | 71.7% | 1 |
| 66 | recraftv3 | 71.6% | 1 |
| 67 | flux11pro | 71.2% | 1 |
| 68 | ideogram20 | 71.0% | 1 |
| 69 | lucidorigin | 71.0% | 1 |
| 70 | glmimage | 70.9% | 1 |
| 71 | gemini20flashpreviewimagegeneration | 68.4% | 1 |
| 72 | flux1devfp8 | 67.9% | 1 |
| 73 | dalle3 | 67.8% | 1 |
| 74 | flux1kontextdev | 66.0% | 1 |
| 75 | stablediffusion35large | 65.7% | 1 |
| 76 | bagel | 62.9% | 1 |
Sistema arena · Confronti a coppie del pubblico
Esplorativo - non è ancora un vincitore in classificarg-image-generation-arena-system-crowd-pairwise-arena-system-v1Evidenze dell'explorer
LMArena Text-to-Image
Fonte unica. Osservato il ; valido fino al .
| Posizione | Configurazione | Capacità | Incertezza | Famiglie |
|---|---|---|---|---|
| 1 | GPT-Image-2.5 Sunburst 2026-10-26config_deca5ad20b2440e13e5172c8e70dd8b61665cced2b19eae98517c8ca25655e50 | 100% | Sconosciuto | 1 |
| 2 | GPT-Image-2.5 Flare 2026-10-26config_1c71330e66d0457bb05baaa848bbbbc58b6e01fbd948a84a4e9416cee082c75a | 95% | Sconosciuto | 1 |
| 3 | GPT-Image-2 (Medium) 2026-10-26config_0e8a3f14213f4bc2a851d65639e1ef46351ff9e25faddc6929cdf5a07c9c4888 | 91% | Sconosciuto | 1 |
| 4 | MAI-Image-2.6 2026-10-26config_7c9ed6221526e113462892af242f8060711cfb6ed526aff8638f38db90c8747a | 81% | Sconosciuto | 1 |
| 5 | Gemini Nano Banana 2.1 2026-10-26config_baafbf6e7f5c67e88d7df8d36baaffacbae3cf471d4f5941a3d401a241bee098 | 81% | Sconosciuto | 1 |
| 6 | Grok Imagine Image 2.0 2026-10-26config_4db0ee8b2fe3c87052d4606468183dcaba9d3abb4a70af75f13e921606301f41 | 79% | Sconosciuto | 1 |
| 7 | Reve 2.1 2026-10-26config_ad38c0bca99e41e885f4c61b980b4c10d251e6fae01f2652b02174532034877d | 76% | Sconosciuto | 1 |
| 8 | Grok Imagine Image 2.0 (20260801) 2026-10-26config_88f25939b7e0c7ab5e246f15dc51f856269b68f1ff157c75ae8b4fda8dd7c6ad | 74% | Sconosciuto | 1 |
| 9 | Muse Image 2026-10-26config_1d15a70c624221bbb20e3e78680eda94b05db1ffc5ea49cb0b3be2d155a80d22 | 70% | Sconosciuto | 1 |
| 10 | Reve 2.0 2026-10-26config_fdb076e63dcc11bcbd65debcdc82859df283dd8e00c3d7edcc27c59fad2c54a6 | 70% | Sconosciuto | 1 |
| 11 | Gemini 3.1 Flash Image (nano-banana-2, Web Search) 2026-10-26config_98294fc76f19b2e6f595b5beeeb0b9e636a6863730a649ce7e30bfc8b99310cb | 68% | Sconosciuto | 1 |
| 12 | Qwen-Image-3.0-Pro 2026-10-26config_66a0cc20632d8fb66882049627b8da631257fe0ff86a91a5553bea955c080be1 | 67% | Sconosciuto | 1 |
| 13 | MAI-Image-2.5 2026-10-26config_3f510e40b8964805ccc7f9e10178a1658f039f3e0d85e1aaeb4baad4989b530f | 67% | Sconosciuto | 1 |
| 14 | Seedream 5.0 Pro 2026-10-26config_f9d0316e988ca730a234e33776e284886b60a94f22fb416fccbaf99c8ef37992 | 67% | Sconosciuto | 1 |
| 15 | Gemini 3.1 Flash-Lite Image (nano-banana-2-lite) 2026-10-26config_e172033c16db32dc1d32a3d6ccc3702efc9717254f018d6b3066be1b3659f264 | 66% | Sconosciuto | 1 |
| 16 | Gemini 3 Pro Image (nano-banana-pro, 2K) 2026-10-26config_8d2b7a8c475eb68e5677cd74c90b6ae76060c01da2c74cdc94d24114c2855e1b | 66% | Sconosciuto | 1 |
| 17 | GPT-Image-1.5 (High Fidelity) 2026-10-26config_b428e20356d90b46814f4f8a286c8495a4280a94ec86532b8fc9c8bf3b8fe71e | 64% | Sconosciuto | 1 |
| 18 | Gemini 3 Pro Image Preview (nano-banana-pro) 2026-10-26config_ada0468e82cb1d31c977f84471ecf515b6cdcdb73a925b286a735fa920ab7693 | 63% | Sconosciuto | 1 |
| 19 | Qwen-Image-2.1 2026-10-26config_164e6b849bb47935764b395b32d14b9bb789aba6ce6a7c526ab1b55c9988d59f | 62% | Sconosciuto | 1 |
| 20 | Ideogram 4.0 (Quality) 2026-10-26config_488246ed6f835f66334c245524d08750c0914035eb495cb9e94c06aea36b35d0 | 57% | Sconosciuto | 1 |
| 21 | Qwen-Image-2.0-Pro (2026-06-22) 2026-10-26config_af3aa52034b5e0fefd853dd61dc90c5a1674737f442b710bb7dc28432d5c4e62 | 55% | Sconosciuto | 1 |
| 22 | UNI-1.1 Max 2026-10-26config_fc8980cfa85f7f756e87a132dba61bbae083e7d15403d7d75f7c252c70646e70 | 55% | Sconosciuto | 1 |
| 23 | MAI-Image-2 2026-10-26config_331b218a0e1a4bd7f8a7ad46a07f5cebaff7fad0257a31df055d93987f086b76 | 54% | Sconosciuto | 1 |
| 24 | UNI-1.1 2026-10-26config_50ae5d75398d0ec4da9d5d8dc372e6b7a9d447b41325c675ccf0ee7c1af44f50 | 54% | Sconosciuto | 1 |
| 25 | Grok Imagine Image 2026-10-26config_1dfe6cad8e01107c5a70882f636c85d651a4d0c4ff6f34f3acfc6f856b7cdfd6 | 51% | Sconosciuto | 1 |
| 26 | Recraft V4.1 Utility Pro 2026-10-26config_7b08a4f49c038a4c3aa7b847d2f635d5f8a7dd8c43c96ffb691d22b091940479 | 51% | Sconosciuto | 1 |
| 27 | Cosmos3-Super-Text2Image (Agentic) 2026-10-26config_80d4b09881eef4b7e5ab4cda18ed830f155247883ca46601b7d4e73f5a835860 | 50% | Sconosciuto | 1 |
| 28 | FLUX.2 [max] 2026-10-26config_fa57870c5af5c728bfa53b458fffa5cad108cf78475a3199dcee8481df0e87ba | 50% | Sconosciuto | 1 |
| 29 | Grok Imagine Image Pro 2026-10-26config_576269b9bae193214dfcf215ad29c9c58784b6b4201c83cd544e84eaa2c659cf | 50% | Sconosciuto | 1 |
| 30 | FLUX.2 [flex] 2026-10-26config_4a021189fad9adc9e51c45adc3d8211bacb76376b0e6edcf3922c54ca10c4c44 | 49% | Sconosciuto | 1 |
| 31 | Reve v1.5 2026-10-26config_8a1a44f7db037df9e533efc083f8f43e3035b7a1f66f0c1f0665ef253f692bc1 | 48% | Sconosciuto | 1 |
| 32 | FLUX.2 [pro] 2026-10-26config_b5d052b8356d136900f6fc1941e35a9ce40eaa6f92d1cd13af61a540b3f3fdbc | 48% | Sconosciuto | 1 |
| 33 | HunyuanImage 3.0 2026-10-26config_e35b1ab7eb08f788d0218edccab00b1c49da37ba066b9c6e583a72fe5f970016 | 48% | Sconosciuto | 1 |
| 34 | Cosmos3-Super-Text2Image 2026-10-26config_692eb089209412757b5248f7bc6d1b16bb3166d61def0f7cd2b07298c7d84670 | 48% | Sconosciuto | 1 |
| 35 | Seedream 4.5 2026-10-26config_f40b65ae8f06c80762103c5d72f2b498ac1797a51592bbf46dcfbf7b44505ca2 | 47% | Sconosciuto | 1 |
| 36 | Gemini 2.5 Flash Image Preview (nano-banana) 2026-10-26config_6db509c76ed444ef8b28b204e61bf1e8869b4c15a40c16d269a7f8dfc73896c4 | 47% | Sconosciuto | 1 |
| 37 | Imagen 4 Ultra (generate-001) 2026-10-26config_3e9b3c8e27d20b30a740eef053fcc9b2bbf4119f4efdf570b7b14b0a0d3ab761 | 47% | Sconosciuto | 1 |
| 38 | FLUX.2 [dev] 2026-10-26config_a7c8166f475d20dd584c765d0119edbf75bb0e1c2d6a5bbd07a0bd4fbaa05981 | 47% | Sconosciuto | 1 |
| 39 | Seedream 5.0 Lite 2026-10-26config_247094ad91dc31945e44a70df83b941a1c49a554eca7d68d3cde527b1129e4b0 | 46% | Sconosciuto | 1 |
| 40 | Seedream 4.0 (2K) 2026-10-26config_beb77706217c91fae5cf365d3217a51db13e188916b0e0ab6663824fb4d8afb7 | 46% | Sconosciuto | 1 |
| 41 | Wan 2.6 2026-10-26config_f0da68773e9e02367394a154def73a730e76cd056cc04ff4fd2adc72acf42232 | 45% | Sconosciuto | 1 |
| 42 | Recraft V4.1 Pro 2026-10-26config_643f6f826d0885941736ba26517b21bc057c57a059845b491e35be3028b4c544 | 44% | Sconosciuto | 1 |
| 43 | Imagen 4 (generate-001) 2026-10-26config_1edc78c5bcdcbfb55dfcd4d92bef529a790500a1f823f48202b983b6569b4d37 | 44% | Sconosciuto | 1 |
| 44 | Krea 2 Medium 2026-10-26config_d42b7242aa5a4a9643a065750b4ef639812229e24aba49d2fc8aa7f5ed483f89 | 44% | Sconosciuto | 1 |
| 45 | Qwen-Image (2512) 2026-10-26config_369a77d9edfbce83a605d5e53e8368f26d494d5111bc23d6f0c4083040322fd1 | 43% | Sconosciuto | 1 |
| 46 | Wan 2.5 Preview 2026-10-26config_f00c97f6261f8bda777d60fb8158ec47f5877aae8dfa6c88e00c53d09bbfe760 | 41% | Sconosciuto | 1 |
| 47 | Seedream 4.0 (fal) 2026-10-26config_84c77336f78dea1bb84d08daca08cdfbc5da15de43190d4a045c850466daf155 | 41% | Sconosciuto | 1 |
| 48 | Krea 2 Large 2026-10-26config_453e7e991c1bc6e166bf62be122dd5751dd30683f412a3ff5fa34eb4fe676fb0 | 41% | Sconosciuto | 1 |
| 49 | GPT-Image-1 2026-10-26config_bf770e4e9af5e8e5fb07093eff7a5783334b35bbab8334fd3cd81ed302909f35 | 41% | Sconosciuto | 1 |
| 50 | Recraft V4 2026-10-26config_fa07ef69835697d23e3b040fb04139581f99440cf5483b27820bc84a80700ac7 | 41% | Sconosciuto | 1 |
| 51 | HiDream-O1-Image 2026-10-26config_5f82c9ef9e863a2506a7a34be84d5b001501e27dbb75ccf009e902eeab04727d | 41% | Sconosciuto | 1 |
| 52 | Krea 2 Turbo 2026-10-26config_678dfa998cceb9f4f50e563ca05721b79f70a2aa488cea8460562b952dc8c026 | 41% | Sconosciuto | 1 |
| 53 | Seedream 4.0 (High Res, fal) 2026-10-26config_afc503f84b07f8906885ed1a776dbad94bc693e370cbab2d100fe008cb9dc4fd | 41% | Sconosciuto | 1 |
| 54 | GPT-Image-1 mini 2026-10-26config_f12277541eda1fca7d4a697f15e5fe8583f4443be796b9fe8c02c0f4073f84f5 | 40% | Sconosciuto | 1 |
| 55 | Wan 2.7 Image Pro 2026-10-26config_772e7a76ae72dc48b7dedfb0f26a6d97edc2aa01da657afa949cd7dd796d73e7 | 39% | Sconosciuto | 1 |
| 56 | Wan 2.7 Image 2026-10-26config_6314d2d07de0b4e501b0d8299b51f085d2d9ebd3aea3551f0eb66aceaca5ffbc | 38% | Sconosciuto | 1 |
| 57 | Recraft V4.1 Flash 2026-10-26config_a80a0da2dd3fb12f8d577fed9244b427e6a04d50aec53ba6cd1d0e446070aca2 | 38% | Sconosciuto | 1 |
| 58 | MAI-Image-1 2026-10-26config_cd66a0f2ae9f4b6abf1bb0ea58266b452768db227cb9fe31a585e3574c1ddf1d | 37% | Sconosciuto | 1 |
| 59 | Z-Image-Turbo 2026-10-26config_6fa17219395079dabcb86351c33bb945b0ef57cc51cb3c1d9893c15e5b0839cb | 35% | Sconosciuto | 1 |
| 60 | Seedream 3.0 2026-10-26config_b01c2e5cc0fa5fd64aa60e40f57eb23bcd841f5409855c5fecd8bfe38bcb6591 | 35% | Sconosciuto | 1 |
| 61 | FLUX.1 Kontext [max] 2026-10-26config_a38d1511cf321e5af8bb03a597e68a92053450a187ca5b0adc792335d00fb8b7 | 34% | Sconosciuto | 1 |
| 62 | FLUX.2 [klein] 9B 2026-10-26config_b94e24775f2504859f7b99d6992a4c8a31096b3aa0c48a45c6bc08cf6d05da0a | 33% | Sconosciuto | 1 |
| 63 | Qwen-Image (Prompt Extend) 2026-10-26config_304ab524a6d2217135e187a92aed226df77078880dc864fff5816b69d1917dcf | 31% | Sconosciuto | 1 |
| 64 | FLUX.1 Kontext [pro] 2026-10-26config_81a26194a1f296cbe9d4a4c267f9a54c28c93f7f0c7a8961a244c23b3dd98fff | 30% | Sconosciuto | 1 |
| 65 | Imagen 3 (generate-002) 2026-10-26config_6b8c85e2c7a19e2b6269c66408cdcbe7fd7377d79051b564ae7633ee9b5d1c9e | 30% | Sconosciuto | 1 |
| 66 | Qwen-Image 2026-10-26config_bb735f9b0ac5f4490d6dd1f0d8aa2a1ef706e04527096e483062b13791e068ac | 30% | Sconosciuto | 1 |
| 67 | Ideogram 3.0 (Quality) 2026-10-26config_7285d4f2fd1ebd1bc0dd835bafcd5c2beffa94c7c7d0453046184d09a05a4e02 | 29% | Sconosciuto | 1 |
| 68 | Luma Photon 2026-10-26config_957b6d8b0cd3107302e3c95f593329967bf3313405281320df16265d47c0ddbd | 26% | Sconosciuto | 1 |
| 69 | P-Image 2026-10-26config_5b55be994a8983dfbe7683d9f22b0032712e9f28bc5b5f79cc62e6371929a2b2 | 25% | Sconosciuto | 1 |
| 70 | FLUX.2 [klein] 4B 2026-10-26config_762c0d9f1b751f0e009e80378d9088a0acb5d83ed1fd8320d00555fd85cd314a | 25% | Sconosciuto | 1 |
| 71 | Runway Gen-4 2026-10-26config_5596795b4757f6d79b0a8c87da0825faf616a4bea882be42756e8740ec5fcc9d | 24% | Sconosciuto | 1 |
| 72 | Recraft V3 2026-10-26config_d8f705610a34e7cd9a7105f54b474cd1297afc31d8447ae53fc0a1111e6f6d05 | 23% | Sconosciuto | 1 |
| 73 | FLUX1.1 [pro] 2026-10-26config_72c57ff1eaafaeacbc6c8d5725efe92ee17afe69b32f0b18e7bef74c6fb4b3d2 | 22% | Sconosciuto | 1 |
| 74 | Ideogram 2.0 2026-10-26config_a05a133e653b18027e2ccf5828082e753b09acce90be59d10e3f7539a277e6ec | 22% | Sconosciuto | 1 |
| 75 | Lucid Origin 2026-10-26config_c699060aed248afc14adbe0e386c5c3aebedd8d9eddc26258ad39212f5217e13 | 22% | Sconosciuto | 1 |
| 76 | GLM-Image 2026-10-26config_9083157cfdf2c365e30b45f24db9906a3794d9b6a0b71393cd28324e483843ce | 22% | Sconosciuto | 1 |
| 77 | Gemini 2.0 Flash Preview Image Generation 2026-10-26config_91bc23e751543a7982fcd8a20b2175aeefd08d744ef61f4ebbd7165d4b9d654c | 15% | Sconosciuto | 1 |
| 78 | FLUX.1 [dev] (FP8) 2026-10-26config_d5fb9a4fd462bffd4c2b9be5c85bfb171f0c45d4e4b64274b0d7671c7889c21a | 14% | Sconosciuto | 1 |
| 79 | DALL-E 3 2026-10-26config_7c134efc4ce6c7240c7f6a5fa440b802e24e3ea524fe1d1db5e0df7298274987 | 13% | Sconosciuto | 1 |
| 80 | FLUX.1 Kontext [dev] 2026-10-26config_cb7d4660803084efe2bec8551192c528e09f9ff4f7dd22a45f8f388fd116389c | 8% | Sconosciuto | 1 |
| 81 | Stable Diffusion 3.5 Large 2026-10-26config_28de4a755649297f7b0080b5b7783866acc8486ab2a19c3bb7aa5cc15fbb4554 | 8% | Sconosciuto | 1 |
| 82 | BAGEL 2026-10-26config_5a61ccf58fda79b4030e347da39fa79b0b2df4716fa2c65befe50d0e91131472 | 0% | Sconosciuto | 1 |
Non ci sono abbastanza evidenze indipendenti per un gruppo di testa
- La calibrazione non è stata convalidata
- La copertura indipendente delle famiglie di benchmark è sotto la soglia richiesta