Рейтинг текстовых моделей
Таблица по открытым замерам оценок людей: оценка, интервал и число голосов; переключайте категорию и дату снимка.
Динамика рейтинга
Загружаем историю замеров…
Как считается
Оценки и интервалы приведены как в открытых замерах оценок людей, без пересчёта; доля использования — по публичным данным API-агрегаторов (топ-K моделей и строка «прочие»); характеристики и цены — из каталогов и провайдеров с датой наблюдения. Интервал у оценки не равен диапазону мест: соседние модели с пересекающимися интервалами статистически неразличимы.
Полная методика: источники, частота обновления, что означает число
Смысл числа
- Метрика
- рейтинг Брэдли — Терри (Elo-подобная шкала) с интервалом из источника
- Единица выборки
- голос в парном сравнении
- Окно
- снимок датасета от 13 сентября 2026
- Как читать
- интервал не равен диапазону мест: модели с пересекающимися интервалами статистически неразличимы
Таблица борда
Снимок от 13 сентября 2026, категория «общий зачёт», 300 строк. Оценка и интервал — из источника; интервал не равен диапазону мест.
| Место | Модель | Разработчик | Оценка | Интервал | Голоса | Лицензия |
|---|---|---|---|---|---|---|
| 1 | Claude Fable 5.1 в режиме max | Anthropic | 1 508 | 1 499–1 516 | 5 783 | Proprietary |
| 2 | Claude Opus 5 в режиме max | Anthropic | 1 505 | 1 500–1 510 | 20 706 | Proprietary |
| 3 | Claude Opus 5 в режиме high | Anthropic | 1 505 | 1 501–1 509 | 42 617 | Proprietary |
| 4 | Claude Opus 4.6 в режиме high | Anthropic | 1 503 | 1 499–1 506 | 71 993 | Proprietary |
| 5 | Claude Opus 4.6 | Anthropic | 1 498 | 1 494–1 501 | 75 878 | Proprietary |
| 6 | Gemini 3.8 Flash в режиме high | 1 495 | 1 486–1 503 | 5 076 | Proprietary | |
| 7 | Claude Fable 5 | Anthropic | 1 493 | 1 488–1 497 | 30 057 | Proprietary |
| 8 | Gemini 3.7 Flash в режиме high | 1 490 | 1 482–1 499 | 5 640 | Proprietary | |
| 9 | Claude Opus 4.7 в режиме high | Anthropic | 1 490 | 1 486–1 494 | 60 002 | Proprietary |
| 10 | Muse Spark 1.3 в режиме max | Meta | 1 490 | 1 481–1 499 | 4 723 | Proprietary |
| 11 | Muse Spark 1.2 в режиме xhigh | Meta | 1 489 | 1 479–1 500 | 3 227 | Proprietary |
| 12 | Claude Opus 4.7 | Anthropic | 1 483 | 1 479–1 487 | 61 128 | Proprietary |
| 13 | Gemini 3.5 Flash в режиме high | 1 482 | 1 478–1 486 | 38 257 | Proprietary | |
| 14 | Qwen3.8 Max (0902) | Alibaba Qwen | 1 481 | 1 475–1 486 | 16 670 | Proprietary |
| 15 | Muse Spark 1.1 | Meta | 1 480 | 1 475–1 485 | 27 615 | Proprietary |
| 16 | Gemini 3.1 Pro Preview | 1 480 | 1 477–1 483 | 106 951 | Proprietary | |
| 17 | Gemini 3 Pro | 1 479 | 1 476–1 483 | 40 654 | Proprietary | |
| 18 | Gemini 3.6 Flash в режиме high | 1 476 | 1 471–1 481 | 26 445 | Proprietary | |
| 19 | Gemini 3.5 Flash в режиме medium | 1 476 | 1 471–1 480 | 36 627 | Proprietary | |
| 20 | GLM 5.3 в режиме max | Z.ai | 1 475 | 1 469–1 481 | 10 960 | MIT |
| 21 | Qwen3.7 Max | Alibaba Qwen | 1 474 | 1 464–1 484 | 3 705 | Proprietary |
| 22 | Muse Spark | Meta | 1 473 | 1 468–1 479 | 13 565 | Proprietary |
| 23 | Kimi K3 в режиме max | Moonshot AI | 1 472 | 1 467–1 477 | 20 987 | Kimi K3 license |
| 24 | GLM 5.3 Flash | Z.ai | 1 472 | 1 465–1 478 | 10 038 | MIT |
| 25 | Qwen3.5 Max | Alibaba Qwen | 1 471 | 1 466–1 476 | 21 476 | Proprietary |
| 26 | GPT-5.5 в режиме high | OpenAI | 1 471 | 1 467–1 475 | 64 924 | Proprietary |
| 27 | GPT-5.4 в режиме high | OpenAI | 1 470 | 1 466–1 473 | 60 537 | Proprietary |
| 28 | ERNIE 5.1 | Baidu | 1 468 | 1 464–1 473 | 37 058 | Proprietary |
| 29 | GLM 5.2 в режиме max | Z.ai | 1 467 | 1 462–1 472 | 36 798 | MIT |
| 30 | Gemini 3 Flash Preview | 1 467 | 1 462–1 471 | 30 225 | Proprietary | |
| 31 | GPT-5.5 · 5.5 | OpenAI | 1 466 | 1 462–1 469 | 66 317 | Proprietary |
| 32 | MiMo-V2.5-Pro | Xiaomi | 1 465 | 1 461–1 469 | 60 919 | MIT |
| 33 | GLM 5.1 | Z.ai | 1 462 | 1 459–1 466 | 48 901 | MIT |
| 34 | Claude Opus 4.8 в режиме high | Anthropic | 1 461 | 1 457–1 465 | 52 535 | Proprietary |
| 35 | Claude Sonnet 4.6 | Anthropic | 1 458 | 1 455–1 462 | 66 208 | Proprietary |
| 36 | Gemini 2.5 Pro | 1 458 | 1 455–1 460 | 122 554 | Proprietary | |
| 37 | GPT-5.6 Sol в режиме xhigh | OpenAI | 1 455 | 1 450–1 460 | 27 069 | Proprietary |
| 38 | Kimi K2.6 | Moonshot AI | 1 455 | 1 450–1 459 | 37 502 | Modified MIT |
| 39 | Qwen3.7 Plus | Alibaba Qwen | 1 454 | 1 450–1 459 | 39 334 | Proprietary |
| 40 | Claude Opus 4.8 | Anthropic | 1 453 | 1 449–1 457 | 53 446 | Proprietary |
| 41 | GPT-5.4 | OpenAI | 1 453 | 1 449–1 456 | 63 526 | Proprietary |
| 42 | Grok 4.20 · beta-0309-reasoning | xAI | 1 451 | 1 447–1 454 | 62 168 | Proprietary |
| 43 | DeepSeek V4 Pro 0423 | DeepSeek | 1 451 | 1 447–1 455 | 54 130 | MIT |
| 44 | Claude Opus 4.5 | Anthropic | 1 450 | 1 447–1 454 | 70 013 | Proprietary |
| 45 | Grok 4.5 | xAI | 1 450 | 1 445–1 455 | 30 103 | Proprietary |
| 46 | Grok 4.20 Multi-Agent | xAI | 1 450 | 1 446–1 454 | 60 777 | Proprietary |
| 47 | Amazon Nova Experimental Chat · 26-02-10 | Amazon | 1 448 | 1 438–1 458 | 3 369 | Proprietary |
| 48 | Seed 2.0 Pro | ByteDance Seed | 1 448 | 1 444–1 451 | 74 159 | Proprietary |
| 49 | Claude Opus 4.5 в режиме high | Anthropic | 1 447 | 1 443–1 451 | 36 239 | Proprietary |
| 50 | Qwen3.6 Max Preview | Alibaba Qwen | 1 446 | 1 438–1 455 | 5 186 | Proprietary |
| 51 | GLM 5 | Z.ai | 1 446 | 1 442–1 451 | 27 605 | MIT |
| 52 | GPT-5.6 Terra в режиме xhigh | OpenAI | 1 446 | 1 441–1 451 | 28 119 | Proprietary |
| 53 | Kimi K2.5 · thinking | Moonshot AI | 1 446 | 1 442–1 449 | 70 513 | Modified MIT |
| 54 | DeepSeek V4 Pro 0423 в режиме high | DeepSeek | 1 445 | 1 441–1 449 | 51 485 | MIT |
| 55 | Nemotron 3 Ultra | NVIDIA | 1 445 | 1 438–1 452 | 10 693 | OpenMDW-1.1 |
| 56 | ERNIE 5.0 · 0110 | Baidu | 1 444 | 1 441–1 448 | 34 860 | Proprietary |
| 57 | Grok 4.20 · beta1 | xAI | 1 444 | 1 440–1 449 | 26 599 | Proprietary |
| 58 | DeepSeek V4 Pro 0813 в режиме high | DeepSeek | 1 444 | 1 437–1 451 | 9 008 | MIT |
| 59 | GPT-6 Astra в режиме max | OpenAI | 1 444 | 1 432–1 455 | 2 693 | Proprietary |
| 60 | ERNIE 5.0 · preview-1203 | Baidu | 1 443 | 1 436–1 449 | 9 561 | Proprietary |
| 61 | Gemini 3 Flash Preview в режиме minimal | 1 442 | 1 439–1 445 | 85 799 | Proprietary | |
| 62 | Claude Sonnet 5 в режиме high | Anthropic | 1 442 | 1 438–1 447 | 35 301 | Proprietary |
| 63 | GPT-5.1 в режиме high | OpenAI | 1 442 | 1 438–1 446 | 40 284 | Proprietary |
| 64 | Gemma 4 31B | 1 442 | 1 434–1 449 | 5 894 | Apache 2.0 | |
| 65 | Hy3 | Tencent | 1 441 | 1 433–1 448 | 8 047 | Apache 2.0 |
| 66 | GLM 4.6 | Z.ai | 1 440 | 1 437–1 444 | 35 061 | MIT |
| 67 | Inkling | Thinking Machines Lab | 1 440 | 1 435–1 445 | 25 922 | Apache 2.0 |
| 68 | Qwen3.8 27B | Alibaba Qwen | 1 439 | 1 433–1 446 | 10 697 | Apache 2.0 |
| 69 | Qwen3 Max · preview | Alibaba Qwen | 1 439 | 1 435–1 444 | 27 194 | Proprietary |
| 70 | GPT-5.2 Chat | OpenAI | 1 439 | 1 435–1 443 | 34 176 | Proprietary |
| 71 | Qwen3.5 397B A17B | Alibaba Qwen | 1 438 | 1 435–1 442 | 77 007 | Apache 2.0 |
| 72 | Claude Sonnet 4.5 | Anthropic | 1 438 | 1 435–1 441 | 79 628 | Proprietary |
| 73 | Grok 4.1 · 4.1-thinking | xAI | 1 437 | 1 434–1 440 | 64 152 | Proprietary |
| 74 | Qwen3.6 Plus | Alibaba Qwen | 1 437 | 1 433–1 441 | 45 319 | Proprietary |
| 75 | MiMo-V2 Pro | Xiaomi | 1 436 | 1 432–1 441 | 24 365 | Proprietary |
| 76 | GLM 5V Turbo | Z.ai | 1 436 | 1 430–1 443 | 9 361 | Proprietary |
| 77 | Grok 4.1 · 4.1 | xAI | 1 436 | 1 433–1 439 | 66 340 | Proprietary |
| 78 | GLM 4.7 | Z.ai | 1 436 | 1 430–1 442 | 11 893 | MIT |
| 79 | Gemini 3.5 Flash Lite | 1 436 | 1 431–1 440 | 26 165 | Proprietary | |
| 80 | Gemma 4 26B A4B | 1 434 | 1 427–1 442 | 5 804 | Apache 2.0 | |
| 81 | Claude Sonnet 4.5 в режиме high | Anthropic | 1 434 | 1 431–1 437 | 80 980 | Proprietary |
| 82 | MiniMax M3 | MiniMax | 1 433 | 1 429–1 438 | 48 540 | MiniMax Community License |
| 83 | DeepSeek V4 Flash 0423 | DeepSeek | 1 432 | 1 428–1 436 | 48 887 | MIT |
| 84 | ERNIE 5.0 · preview-1022 | Baidu | 1 430 | 1 422–1 439 | 4 663 | Proprietary |
| 85 | GLM 4.5 | Z.ai | 1 430 | 1 425–1 435 | 23 712 | MIT |
| 86 | GPT-5.6 Luna в режиме xhigh | OpenAI | 1 430 | 1 425–1 435 | 28 547 | Proprietary |
| 87 | Grok 4.6 в режиме high | xAI | 1 430 | 1 424–1 436 | 15 521 | Proprietary |
| 88 | ChatGPT-4o (latest) | OpenAI | 1 429 | 1 426–1 432 | 80 677 | Proprietary |
| 89 | R1 0528 | DeepSeek | 1 428 | 1 422–1 433 | 18 091 | MIT |
| 90 | MiMo-V2.5 | Xiaomi | 1 427 | 1 423–1 432 | 44 466 | MIT |
| 91 | Mistral Large 3 2512 | Mistral AI | 1 427 | 1 424–1 430 | 69 028 | Apache 2.0 |
| 92 | LongCat-Flash-Chat · chat-2602-exp | Meituan | 1 426 | 1 421–1 430 | 27 962 | Proprietary |
| 93 | Grok 3 | xAI | 1 425 | 1 421–1 430 | 32 414 | Proprietary |
| 94 | DeepSeek V3.2 Exp · exp-thinking | DeepSeek | 1 425 | 1 418–1 432 | 8 905 | MIT |
| 95 | Mistral Medium 3.1 | Mistral AI | 1 425 | 1 422–1 427 | 92 697 | Proprietary |
| 96 | DeepSeek V3.2 · v3.2 | DeepSeek | 1 425 | 1 421–1 428 | 46 458 | MIT |
| 97 | DeepSeek V4 Flash 0423 в режиме high | DeepSeek | 1 424 | 1 420–1 428 | 48 597 | MIT |
| 98 | LongCat-Flash-Chat · chat | Meituan | 1 423 | 1 417–1 429 | 11 154 | MIT |
| 99 | DeepSeek V3.2 Exp · exp | DeepSeek | 1 423 | 1 416–1 429 | 11 760 | MIT |
| 100 | GPT-5.1 | OpenAI | 1 423 | 1 419–1 426 | 42 982 | Proprietary |
| 101 | MiMo-V2 Omni | Xiaomi | 1 422 | 1 417–1 428 | 19 414 | Proprietary |
| 102 | Amazon Nova Experimental Chat · 12-10 | Amazon | 1 421 | 1 412–1 431 | 3 632 | Proprietary |
| 103 | Qwen3 VL 235B A22B Instruct | Alibaba Qwen | 1 421 | 1 414–1 427 | 11 219 | Apache 2.0 |
| 104 | Mistral Medium 3.5 | Mistral AI | 1 421 | 1 414–1 427 | 10 996 | Modified MIT |
| 105 | Kimi K2.5 · instant | Moonshot AI | 1 420 | 1 414–1 427 | 7 987 | Modified MIT |
| 106 | DeepSeek V3.1 Terminus · terminus-thinking | DeepSeek | 1 420 | 1 410–1 430 | 3 369 | MIT |
| 107 | DeepSeek V3.2 · v3.2-thinking | DeepSeek | 1 420 | 1 416–1 423 | 40 499 | MIT |
| 108 | Qwen3 235B A22B Instruct 2507 | Alibaba Qwen | 1 420 | 1 417–1 422 | 95 670 | Apache 2.0 |
| 109 | GPT-5.5 · 5.5-instant | OpenAI | 1 420 | 1 415–1 425 | 25 850 | Proprietary |
| 110 | DeepSeek V3.1 · v3.1 | DeepSeek | 1 420 | 1 414–1 426 | 14 607 | MIT |
| 111 | Claude Opus 4.1 · 20250805-thinking-16k | Anthropic | 1 419 | 1 416–1 423 | 48 811 | Proprietary |
| 112 | Claude Opus 4.1 · 20250805 | Anthropic | 1 418 | 1 415–1 421 | 75 913 | Proprietary |
| 113 | Qwen3.5-122B-A10B | Alibaba Qwen | 1 418 | 1 414–1 422 | 28 359 | Apache 2.0 |
| 114 | Qwen3 Next 80B A3B Instruct | Alibaba Qwen | 1 418 | 1 413–1 422 | 22 529 | Apache 2.0 |
| 115 | GPT-4.5 Preview | OpenAI | 1 417 | 1 412–1 423 | 14 547 | Proprietary |
| 116 | Gemini 2.5 Flash · flash | 1 417 | 1 415–1 420 | 122 732 | Proprietary | |
| 117 | DeepSeek V3.1 Terminus · terminus | DeepSeek | 1 417 | 1 407–1 427 | 3 611 | MIT |
| 118 | DeepSeek V3.1 · v3.1-thinking | DeepSeek | 1 417 | 1 410–1 423 | 11 460 | MIT |
| 119 | GPT-5.2 в режиме high | OpenAI | 1 416 | 1 413–1 420 | 47 538 | Proprietary |
| 120 | Amazon Nova Experimental Chat · 11-10 | Amazon | 1 416 | 1 412–1 420 | 25 086 | Proprietary |
| 121 | Gemini 3.1 Flash Lite Preview | 1 415 | 1 412–1 419 | 60 405 | Proprietary | |
| 122 | Qwen3 235B A22B Thinking 2507 | Alibaba Qwen | 1 415 | 1 409–1 422 | 8 796 | Apache 2.0 |
| 123 | Kimi K2 Thinking | Moonshot AI | 1 414 | 1 411–1 418 | 61 088 | Modified MIT |
| 124 | Qwen3 Max · 2025-09-23 | Alibaba Qwen | 1 413 | 1 407–1 420 | 8 939 | Proprietary |
| 125 | GPT-5.2 | OpenAI | 1 412 | 1 409–1 416 | 78 967 | Proprietary |
| 126 | Inkling Small | Thinking Machines Lab | 1 412 | 1 407–1 418 | 18 844 | Apache 2.0 |
| 127 | GPT-5.4 Mini в режиме high | OpenAI | 1 412 | 1 408–1 416 | 59 387 | Proprietary |
| 128 | Grok 4 | xAI | 1 412 | 1 408–1 415 | 39 297 | Proprietary |
| 129 | MiMo-V2 Flash · flash (non-thinking) | Xiaomi | 1 411 | 1 408–1 415 | 46 116 | MIT |
| 130 | o3 | OpenAI | 1 410 | 1 406–1 413 | 58 579 | Proprietary |
| 131 | Qwen3.5-27B | Alibaba Qwen | 1 408 | 1 404–1 412 | 27 227 | Apache 2.0 |
| 132 | Grok 4.1 Fast | xAI | 1 408 | 1 405–1 411 | 55 363 | Proprietary |
| 133 | Gemini 2.5 Flash · flash-preview-09-2025 | 1 407 | 1 403–1 411 | 32 364 | Proprietary | |
| 134 | Grok 4 Fast · chat | xAI | 1 407 | 1 399–1 414 | 6 335 | Proprietary |
| 135 | Hunyuan Vision 1.5 | Tencent | 1 406 | 1 394–1 419 | 2 171 | Proprietary |
| 136 | GPT-5 в режиме high | OpenAI | 1 406 | 1 402–1 411 | 31 428 | Proprietary |
| 137 | Hy3 preview | Tencent | 1 405 | 1 397–1 413 | 6 614 | tencent-hunyuan-community |
| 138 | MiniMax M2.7 | MiniMax | 1 405 | 1 401–1 408 | 69 169 | Modified MIT |
| 139 | GPT-5 | OpenAI | 1 404 | 1 399–1 408 | 30 957 | Proprietary |
| 140 | Step 3.5 Flash | StepFun | 1 404 | 1 400–1 407 | 57 137 | Apache 2.0 |
| 141 | Qwen3 VL 235B A22B Thinking | Alibaba Qwen | 1 401 | 1 394–1 408 | 7 780 | Apache 2.0 |
| 142 | Hunyuan T1 | Tencent | 1 401 | 1 392–1 409 | 4 580 | Proprietary |
| 143 | Grok 4 Fast · reasoning | xAI | 1 399 | 1 394–1 404 | 17 975 | Proprietary |
| 144 | Amazon Nova Experimental Chat · 26-01-10 | Amazon | 1 398 | 1 388–1 408 | 3 346 | Proprietary |
| 145 | Grok 4.3 | xAI | 1 398 | 1 394–1 401 | 66 801 | Proprietary |
| 146 | Qwen3.5-Flash | Alibaba Qwen | 1 398 | 1 394–1 401 | 58 193 | Proprietary |
| 147 | Claude Haiku 4.5 | Anthropic | 1 397 | 1 394–1 399 | 129 278 | Proprietary |
| 148 | Qwen3.5-35B-A3B | Alibaba Qwen | 1 395 | 1 391–1 400 | 29 043 | Apache 2.0 |
| 149 | Amazon Nova Experimental Chat · 10-20 | Amazon | 1 395 | 1 389–1 401 | 11 368 | Proprietary |
| 150 | MiMo-V2 Flash · flash (thinking) | Xiaomi | 1 394 | 1 388–1 401 | 10 739 | MIT |
| 151 | Qwen3 235B A22B · a22b-no-thinking | Alibaba Qwen | 1 394 | 1 389–1 398 | 37 389 | Apache 2.0 |
| 152 | Muse Glimmer 30B | Meta | 1 392 | 1 382–1 402 | 3 666 | Apache-2.0 |
| 153 | MiniMax M2.1 | MiniMax | 1 391 | 1 386–1 396 | 16 622 | MIT |
| 154 | GPT-5.3 Chat | OpenAI | 1 388 | 1 384–1 392 | 32 727 | Proprietary |
| 155 | GLM 4.5 Air | Z.ai | 1 384 | 1 379–1 388 | 30 367 | MIT |
| 156 | Qwen3 30B A3B Instruct 2507 | Alibaba Qwen | 1 384 | 1 379–1 389 | 23 222 | Apache 2.0 |
| 157 | GPT-4.1 | OpenAI | 1 383 | 1 379–1 386 | 49 941 | Proprietary |
| 158 | Kimi K2 0905 | Moonshot AI | 1 380 | 1 373–1 386 | 11 568 | Modified MIT |
| 159 | Gemini 2.5 Flash Lite · 09-2025-no-thinking | 1 379 | 1 376–1 383 | 46 382 | Proprietary | |
| 160 | Nemotron 3 Super | NVIDIA | 1 378 | 1 371–1 385 | 7 458 | NVIDIA Open Model |
| 161 | Solar Pro 4 | Upstage | 1 377 | 1 365–1 389 | 2 420 | Proprietary |
| 162 | Claude Opus 4 · 20250514-thinking-16k | Anthropic | 1 377 | 1 372–1 381 | 35 735 | Proprietary |
| 163 | GLM 4.6V | Z.ai | 1 377 | 1 365–1 388 | 2 762 | MIT |
| 164 | Hunyuan TurboS · 20250416 | Tencent | 1 376 | 1 370–1 383 | 10 586 | Proprietary |
| 165 | DeepSeek V3 0324 | DeepSeek | 1 375 | 1 371–1 379 | 44 787 | MIT |
| 166 | GPT-5 Mini в режиме high | OpenAI | 1 373 | 1 368–1 378 | 26 569 | Proprietary |
| 167 | GPT-5.4 Nano в режиме high | OpenAI | 1 373 | 1 369–1 377 | 58 424 | Proprietary |
| 168 | R1 | DeepSeek | 1 373 | 1 368–1 377 | 18 524 | MIT |
| 169 | Kimi K2 0711 | Moonshot AI | 1 371 | 1 366–1 376 | 27 029 | Modified MIT |
| 170 | Mistral Medium 3 | Mistral AI | 1 370 | 1 365–1 375 | 32 662 | Proprietary |
| 171 | Gemini 2.5 Flash Lite · 06-17-thinking | 1 369 | 1 365–1 374 | 32 216 | Proprietary | |
| 172 | Grok 3 Mini в режиме high | xAI | 1 369 | 1 364–1 375 | 15 536 | Proprietary |
| 173 | Qwen3 Next 80B A3B Thinking | Alibaba Qwen | 1 368 | 1 362–1 374 | 13 380 | Apache 2.0 |
| 174 | Qwen2.5 Max | Alibaba Qwen | 1 367 | 1 362–1 371 | 32 417 | Proprietary |
| 175 | Qwen3 235B A22B · a22b | Alibaba Qwen | 1 366 | 1 361–1 371 | 25 865 | Apache 2.0 |
| 176 | gpt-oss-120b | OpenAI | 1 366 | 1 361–1 370 | 29 955 | Apache 2.0 |
| 177 | o1 · 2024-12-17 | OpenAI | 1 366 | 1 361–1 370 | 27 807 | Proprietary |
| 178 | Claude Opus 4 · 20250514 | Anthropic | 1 366 | 1 361–1 370 | 42 845 | Proprietary |
| 179 | Amazon Nova Experimental Chat · 10-09 | Amazon | 1 364 | 1 353–1 375 | 2 775 | Proprietary |
| 180 | Grok 3 Mini | xAI | 1 364 | 1 359–1 369 | 20 837 | Proprietary |
| 181 | Granite 4.2 30B | IBM Granite | 1 363 | 1 353–1 374 | 3 255 | Apache 2.0 |
| 182 | Nova 2 Lite | Amazon | 1 363 | 1 357–1 369 | 12 120 | Proprietary |
| 183 | Ling-flash-2.0 | inclusionAI | 1 361 | 1 354–1 369 | 6 850 | MIT |
| 184 | MiniMax M2.5 | MiniMax | 1 359 | 1 355–1 363 | 40 843 | Modified MIT |
| 185 | Gemma 3 27B | 1 358 | 1 354–1 361 | 46 422 | Gemma | |
| 186 | Mercury 2 | Inception Labs | 1 358 | 1 347–1 368 | 3 042 | Proprietary |
| 187 | Qwen3 Coder 480B A35B | Alibaba Qwen | 1 356 | 1 351–1 361 | 25 089 | Apache 2.0 |
| 188 | INTELLECT-3 | Prime Intellect | 1 356 | 1 348–1 364 | 5 279 | MIT |
| 189 | Gemini 2.0 Flash | 1 354 | 1 350–1 358 | 43 351 | Proprietary | |
| 190 | o4 Mini | OpenAI | 1 353 | 1 349–1 357 | 44 639 | Proprietary |
| 191 | o1 · preview | OpenAI | 1 353 | 1 348–1 358 | 31 122 | Proprietary |
| 192 | GLM 4.7 Flash | Z.ai | 1 352 | 1 346–1 358 | 11 491 | MIT |
| 193 | Step 3 | StepFun | 1 350 | 1 343–1 358 | 6 273 | Apache 2.0 |
| 194 | Claude Sonnet 4 · 20250514-thinking-32k | Anthropic | 1 350 | 1 346–1 355 | 33 944 | Proprietary |
| 195 | Nemotron 3 Nano 30B A3B | NVIDIA | 1 348 | 1 343–1 354 | 15 325 | NVIDIA Open Model |
| 196 | MiniMax M1 | MiniMax | 1 343 | 1 338–1 347 | 34 030 | Apache 2.0 |
| 197 | MiniMax M2 | MiniMax | 1 342 | 1 334–1 350 | 6 753 | Apache 2.0 |
| 198 | Trinity Large Thinking | Arcee AI | 1 342 | 1 337–1 347 | 28 967 | Apache 2.0 |
| 199 | GPT-4.1 Mini | OpenAI | 1 340 | 1 336–1 345 | 38 631 | Proprietary |
| 200 | Qwen3 32B | Alibaba Qwen | 1 340 | 1 331–1 349 | 3 926 | Apache 2.0 |
| 201 | Claude Sonnet 4 · 20250514 | Anthropic | 1 339 | 1 335–1 344 | 38 966 | Proprietary |
| 202 | Trinity Large | Arcee AI | 1 338 | 1 334–1 343 | 29 742 | Apache 2.0 |
| 203 | Mistral Small 3.2 24B | Mistral AI | 1 338 | 1 333–1 344 | 17 391 | Apache 2.0 |
| 204 | Llama 3.3 Nemotron Super 49B v1.5 | NVIDIA | 1 338 | 1 328–1 348 | 3 313 | Nvidia Open |
| 205 | o3 Mini High | OpenAI | 1 337 | 1 331–1 342 | 18 589 | Proprietary |
| 206 | Step-1o Turbo | StepFun | 1 336 | 1 329–1 343 | 8 858 | Proprietary |
| 207 | Gemma 3 12B | 1 334 | 1 325–1 344 | 3 829 | Gemma | |
| 208 | GLM 4.5V | Z.ai | 1 333 | 1 325–1 341 | 4 817 | MIT |
| 209 | DeepSeek V3 | DeepSeek | 1 332 | 1 328–1 337 | 21 770 | DeepSeek |
| 210 | Ring-flash-2.0 | inclusionAI | 1 332 | 1 325–1 339 | 6 899 | MIT |
| 211 | Command A | Cohere | 1 331 | 1 327–1 334 | 55 449 | CC-BY-NC-4.0 |
| 212 | GLM-4-Plus · plus-0111 | Z.ai | 1 331 | 1 322–1 339 | 5 760 | Proprietary |
| 213 | Nemotron 3.5 Lightning | NVIDIA | 1 331 | 1 323–1 338 | 8 338 | OpenMDW-1.1 |
| 214 | Gemini 2.0 Flash-Lite | 1 329 | 1 325–1 334 | 24 955 | Proprietary | |
| 215 | QwQ 32B | Alibaba Qwen | 1 329 | 1 325–1 334 | 25 090 | Apache 2.0 |
| 216 | Qwen-Plus | Alibaba Qwen | 1 326 | 1 318–1 335 | 5 819 | Proprietary |
| 217 | Step-2 16k | StepFun | 1 321 | 1 312–1 329 | 4 833 | Proprietary |
| 218 | GPT-5 Nano в режиме high | OpenAI | 1 320 | 1 313–1 327 | 8 128 | Proprietary |
| 219 | Hunyuan TurboS · 20250226 | Tencent | 1 320 | 1 308–1 331 | 2 220 | Proprietary |
| 220 | Llama 3.1 Nemotron Ultra 253B v1 | NVIDIA | 1 319 | 1 308–1 331 | 2 549 | Nvidia Open Model |
| 221 | o3 Mini | OpenAI | 1 319 | 1 316–1 323 | 56 655 | Proprietary |
| 222 | Gemini 1.5 Pro · 1.5-pro-002 | 1 319 | 1 316–1 322 | 55 606 | Proprietary | |
| 223 | o1-mini | OpenAI | 1 317 | 1 313–1 321 | 51 981 | Proprietary |
| 224 | Qwen3 30B A3B | Alibaba Qwen | 1 317 | 1 312–1 322 | 26 089 | Apache 2.0 |
| 225 | Granite 4.2 8B | IBM Granite | 1 317 | 1 305–1 328 | 3 087 | Apache 2.0 |
| 226 | Claude 3.7 Sonnet · 20250219-thinking-32k | Anthropic | 1 314 | 1 310–1 319 | 38 081 | Proprietary |
| 227 | OLMo 3.1 32B Instruct | Ai2 | 1 312 | 1 306–1 318 | 11 452 | Apache 2.0 |
| 228 | Hunyuan Turbo | Tencent | 1 311 | 1 300–1 323 | 2 290 | Proprietary |
| 229 | Llama 3.3 Nemotron Super 49B v1 | NVIDIA | 1 308 | 1 296–1 320 | 2 218 | Nvidia |
| 230 | Gemma 3n E4B | 1 305 | 1 300–1 310 | 22 082 | Gemma | |
| 231 | Grok 2 | xAI | 1 304 | 1 301–1 308 | 63 498 | Proprietary |
| 232 | Yi-Lightning | 01.AI | 1 301 | 1 297–1 306 | 27 332 | Proprietary |
| 233 | GPT-4o (2024-05-13) | OpenAI | 1 300 | 1 297–1 304 | 112 881 | Proprietary |
| 234 | OLMo 3 32B Think | Ai2 | 1 300 | 1 291–1 308 | 5 682 | Apache 2.0 |
| 235 | Claude 3.7 Sonnet · 20250219 | Anthropic | 1 299 | 1 295–1 303 | 42 448 | Proprietary |
| 236 | Qwen2.5 Plus | Alibaba Qwen | 1 299 | 1 293–1 305 | 10 187 | Proprietary |
| 237 | Claude 3.5 Sonnet · 20241022 | Anthropic | 1 298 | 1 295–1 301 | 87 694 | Proprietary |
| 238 | Granite 4.2 3B | IBM Granite | 1 297 | 1 286–1 309 | 3 072 | Apache 2.0 |
| 239 | Molmo 2 8B | Ai2 | 1 294 | 1 273–1 315 | 791 | Apache 2.0 |
| 240 | DeepSeek V2.5 · v2.5-1210 | DeepSeek | 1 294 | 1 286–1 302 | 6 795 | DeepSeek |
| 241 | Granite 4.1 8B | IBM Granite | 1 292 | 1 282–1 302 | 4 023 | Apache 2.0 |
| 242 | Athene V2 Chat | Nexusflow | 1 291 | 1 287–1 296 | 24 739 | NexusFlow |
| 243 | Gemma 3 4B | 1 291 | 1 282–1 300 | 4 171 | Gemma | |
| 244 | GLM-4-Plus · plus | Z.ai | 1 290 | 1 285–1 294 | 26 126 | Proprietary |
| 245 | Hunyuan Large | Tencent | 1 288 | 1 278–1 298 | 3 738 | Proprietary |
| 246 | gpt-oss-20b | OpenAI | 1 287 | 1 281–1 294 | 10 393 | Apache 2.0 |
| 247 | Llama 4 Maverick | Meta | 1 287 | 1 283–1 291 | 39 359 | Llama 4 |
| 248 | Gemini 1.5 Flash · 002 | 1 287 | 1 282–1 291 | 34 902 | Proprietary | |
| 249 | GPT-4o-mini (2024-07-18) | OpenAI | 1 286 | 1 283–1 290 | 68 697 | Proprietary |
| 250 | GPT-4.1 Nano | OpenAI | 1 285 | 1 277–1 292 | 6 103 | Proprietary |
| 251 | Mercury | Inception Labs | 1 284 | 1 270–1 298 | 1 930 | Proprietary |
| 252 | Llama 3.1 405B Instruct · bf16 | Meta | 1 284 | 1 280–1 287 | 41 375 | Llama 3.1 Community |
| 253 | Llama 3.1 Nemotron 70B Instruct | NVIDIA | 1 283 | 1 275–1 290 | 7 140 | Llama 3.1 |
| 254 | GPT-4o (2024-08-06) | OpenAI | 1 282 | 1 278–1 287 | 45 499 | Proprietary |
| 255 | Qwen Max | Alibaba Qwen | 1 282 | 1 276–1 287 | 16 478 | Qwen |
| 256 | Llama 3.1 405B Instruct · fp8 | Meta | 1 282 | 1 278–1 285 | 59 656 | Llama 3.1 Community |
| 257 | Claude 3.5 Sonnet · 20240620 | Anthropic | 1 281 | 1 277–1 284 | 82 419 | Proprietary |
| 258 | Grok 2 Mini | xAI | 1 281 | 1 277–1 284 | 52 567 | Proprietary |
| 259 | Llama 4 Scout | Meta | 1 279 | 1 275–1 284 | 29 740 | Llama |
| 260 | Gemini 1.5 Pro · advanced-0514 | 1 278 | 1 273–1 284 | 50 148 | Proprietary | |
| 261 | Mistral Small 3.1 24B | Mistral AI | 1 277 | 1 273–1 282 | 32 618 | Apache 2.0 |
| 262 | Llama 3.3 70B Instruct | Meta | 1 274 | 1 270–1 277 | 54 412 | Llama-3.3 |
| 263 | Hunyuan Standard | Tencent | 1 274 | 1 264–1 284 | 3 904 | Proprietary |
| 264 | Gemini 1.5 Pro · 1.5-pro-001 | 1 273 | 1 269–1 277 | 79 138 | Proprietary | |
| 265 | GPT-4 Turbo | OpenAI | 1 272 | 1 268–1 275 | 98 114 | Proprietary |
| 266 | DeepSeek V2.5 · v2.5 | DeepSeek | 1 271 | 1 266–1 276 | 24 572 | DeepSeek |
| 267 | OLMo 3.1 32B Think | Ai2 | 1 270 | 1 263–1 278 | 8 117 | Apache 2.0 |
| 268 | Qwen2.5 72B Instruct | Alibaba Qwen | 1 269 | 1 265–1 273 | 39 406 | Qwen |
| 269 | Mistral Large 2407 | Mistral AI | 1 266 | 1 262–1 270 | 45 459 | Mistral Research |
| 270 | Mistral Large | Mistral AI | 1 265 | 1 261–1 270 | 28 073 | MRL |
| 271 | Athene 70B | Nexusflow | 1 265 | 1 259–1 271 | 19 621 | CC-BY-NC-4.0 |
| 272 | GPT-4 Turbo Preview · 1106-preview | OpenAI | 1 263 | 1 259–1 267 | 100 105 | Proprietary |
| 273 | Hunyuan Large Vision | Tencent | 1 263 | 1 254–1 272 | 5 259 | Proprietary |
| 274 | GPT-4 Turbo Preview · 0125-preview | OpenAI | 1 262 | 1 258–1 266 | 93 439 | Proprietary |
| 275 | Claude 3 Opus | Anthropic | 1 262 | 1 259–1 265 | 194 909 | Proprietary |
| 276 | Llama 3.1 70B Instruct | Meta | 1 261 | 1 257–1 265 | 55 240 | Llama 3.1 Community |
| 277 | Nova Pro 1.0 | Amazon | 1 259 | 1 254–1 263 | 24 745 | Proprietary |
| 278 | Llama 3.1 Tulu 3 70B | Ai2 | 1 256 | 1 245–1 266 | 2 846 | Llama 3.1 |
| 279 | Claude 3.5 Haiku | Anthropic | 1 255 | 1 252–1 258 | 68 964 | Proprietary |
| 280 | Magistral Medium | Mistral AI | 1 255 | 1 248–1 261 | 11 468 | Proprietary |
| 281 | Reka Core | Reka AI | 1 248 | 1 241–1 255 | 7 312 | Proprietary |
| 282 | Granite 4.0 H Small | IBM Granite | 1 240 | 1 231–1 248 | 5 532 | Apache 2.0 |
| 283 | Gemini 1.5 Flash · 001 | 1 239 | 1 235–1 244 | 62 833 | Proprietary | |
| 284 | Jamba 1.5 Large | AI21 Labs | 1 237 | 1 230–1 244 | 8 662 | Jamba Open |
| 285 | Mistral Small 3 | Mistral AI | 1 233 | 1 228–1 239 | 14 681 | Apache 2.0 |
| 286 | Gemma 2 27B | 1 231 | 1 228–1 235 | 75 754 | Gemma license | |
| 287 | Qwen2.5 Coder 32B Instruct | Alibaba Qwen | 1 230 | 1 222–1 238 | 5 432 | Apache 2.0 |
| 288 | Command R+ (08-2024) | Cohere | 1 229 | 1 222–1 235 | 9 866 | CC-BY-NC-4.0 |
| 289 | Nova Lite 1.0 | Amazon | 1 228 | 1 223–1 234 | 19 372 | Proprietary |
| 290 | Llama 3.1 Nemotron 51B Instruct | NVIDIA | 1 228 | 1 218–1 238 | 3 749 | Llama 3.1 |
| 291 | Gemma 2 9B IT SimPO | Princeton NLP | 1 227 | 1 220–1 234 | 10 072 | MIT |
| 292 | GLM-4 | Z.ai | 1 226 | 1 219–1 233 | 9 788 | Proprietary |
| 293 | Gemini 1.5 Flash-8B | 1 226 | 1 221–1 230 | 35 558 | Proprietary | |
| 294 | Nemotron-4 340B Instruct | NVIDIA | 1 225 | 1 219–1 230 | 19 659 | NVIDIA Open Model |
| 295 | Aya Expanse 32B | Cohere | 1 224 | 1 219–1 229 | 27 124 | CC-BY-NC-4.0 |
| 296 | Llama 3 70B Instruct | Meta | 1 221 | 1 217–1 224 | 156 876 | Llama 3 Community |
| 297 | Claude 3 Sonnet | Anthropic | 1 218 | 1 214–1 222 | 109 284 | Proprietary |
| 298 | OLMo 2 32B Instruct | Ai2 | 1 218 | 1 207–1 229 | 3 334 | Apache-2.0 |
| 299 | Reka Flash (2024-09) | Reka AI | 1 218 | 1 211–1 225 | 7 536 | Proprietary |
| 300 | Phi 4 | Microsoft | 1 217 | 1 212–1 221 | 24 126 | MIT |
Данные на 13 сентября 2026. Как мы считаем: методика
Свежесть данных
Снимок замеров опубликован 13 сентября 2026; всего снимков борда в архиве: 200.
- замеры оценок людей: загружено 18 сентября 2026, 08:25 UTC
- каталог моделей и цены: загружено 18 сентября 2026, 08:51 UTC
- доли использования: загружено 18 сентября 2026, 08:51 UTC
Другие рейтинги
- Рейтинг агентов
- Рейтинг работы с документами
- Рейтинг редактирования изображений
- Рейтинг видео из изображения
- Популярность моделей по использованию через API-агрегаторы
- Рейтинг поисковых моделей
- Рейтинг моделей для кода
- Рейтинг моделей для творческих текстов
- Рейтинг моделей для математики
- Рейтинг моделей для русского языка
- Рейтинг генерации изображений
- Рейтинг генерации видео
- Рейтинг редактирования видео
- Рейтинг моделей с изображениями на входе
- Рейтинг моделей для веб-разработки
Данные и источники
Обновлено 18 сентября 2026. Каталог, цены и режимы проверяются ежедневно по нескольким API-агрегаторам и провайдерам. Места — по открытым замерам оценок людей, интервалы приведены как в замере.