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 «отрасль: развлечения, спорт, медиа», 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 | 1 495 | 1 479–1 512 | 1 426 | Proprietary |
| 2 | Claude Opus 4.6 в режиме high | Anthropic | 1 494 | 1 487–1 500 | 15 786 | Proprietary |
| 3 | Claude Fable 5 | Anthropic | 1 485 | 1 477–1 493 | 7 578 | Proprietary |
| 4 | Claude Opus 4.6 | Anthropic | 1 481 | 1 475–1 487 | 16 342 | Proprietary |
| 5 | Claude Opus 5 в режиме high | Anthropic | 1 479 | 1 471–1 486 | 11 129 | Proprietary |
| 6 | Claude Opus 5 в режиме max | Anthropic | 1 478 | 1 469–1 487 | 5 423 | Proprietary |
| 7 | Gemini 3.7 Flash в режиме high | 1 477 | 1 461–1 493 | 1 570 | Proprietary | |
| 8 | Claude Opus 4.7 в режиме high | Anthropic | 1 472 | 1 465–1 478 | 13 578 | Proprietary |
| 9 | Gemini 3.8 Flash в режиме high | 1 467 | 1 451–1 484 | 1 377 | Proprietary | |
| 10 | Claude Opus 4.7 | Anthropic | 1 466 | 1 460–1 473 | 14 041 | Proprietary |
| 11 | Gemini 3.1 Pro Preview | 1 464 | 1 458–1 469 | 23 874 | Proprietary | |
| 12 | Gemini 3 Pro | 1 463 | 1 456–1 471 | 7 596 | Proprietary | |
| 13 | Qwen3.8 Max (0902) | Alibaba Qwen | 1 461 | 1 451–1 471 | 4 108 | Proprietary |
| 14 | Muse Spark 1.3 в режиме max | Meta | 1 461 | 1 442–1 480 | 1 071 | Proprietary |
| 15 | Gemini 3.5 Flash в режиме high | 1 460 | 1 453–1 467 | 9 115 | Proprietary | |
| 16 | Qwen3.5 Max | Alibaba Qwen | 1 457 | 1 448–1 467 | 4 025 | Proprietary |
| 17 | Muse Spark 1.2 в режиме xhigh | Meta | 1 456 | 1 434–1 478 | 756 | Proprietary |
| 18 | Gemini 3.6 Flash в режиме high | 1 456 | 1 447–1 464 | 6 716 | Proprietary | |
| 19 | Gemini 3.5 Flash в режиме medium | 1 455 | 1 447–1 462 | 9 121 | Proprietary | |
| 20 | GLM 5.3 в режиме max | Z.ai | 1 453 | 1 441–1 465 | 2 923 | MIT |
| 21 | GPT-5.6 Sol в режиме xhigh | OpenAI | 1 450 | 1 441–1 458 | 6 694 | Proprietary |
| 22 | Muse Spark | Meta | 1 449 | 1 437–1 462 | 2 510 | Proprietary |
| 23 | Gemini 3 Flash Preview | 1 449 | 1 441–1 458 | 5 520 | Proprietary | |
| 24 | GPT-5.5 в режиме high | OpenAI | 1 449 | 1 443–1 455 | 15 232 | Proprietary |
| 25 | Qwen3.7 Max | Alibaba Qwen | 1 446 | 1 423–1 469 | 681 | Proprietary |
| 26 | GLM 5.2 в режиме max | Z.ai | 1 445 | 1 438–1 453 | 8 941 | MIT |
| 27 | GPT-5.5 · 5.5 | OpenAI | 1 445 | 1 439–1 451 | 15 678 | Proprietary |
| 28 | GLM 5.1 | Z.ai | 1 443 | 1 437–1 450 | 11 227 | MIT |
| 29 | Muse Spark 1.1 | Meta | 1 443 | 1 435–1 451 | 6 857 | Proprietary |
| 30 | Kimi K3 в режиме max | Moonshot AI | 1 443 | 1 434–1 452 | 5 076 | Kimi K3 license |
| 31 | Gemini 2.5 Pro | 1 440 | 1 436–1 445 | 22 658 | Proprietary | |
| 32 | Claude Opus 4.8 в режиме high | Anthropic | 1 440 | 1 433–1 446 | 12 790 | Proprietary |
| 33 | GPT-6 Astra в режиме max | OpenAI | 1 439 | 1 415–1 463 | 686 | Proprietary |
| 34 | MiMo-V2.5-Pro | Xiaomi | 1 439 | 1 433–1 445 | 13 809 | MIT |
| 35 | GPT-5.4 в режиме high | OpenAI | 1 439 | 1 432–1 445 | 12 916 | Proprietary |
| 36 | GLM 5.3 Flash | Z.ai | 1 438 | 1 425–1 450 | 2 529 | MIT |
| 37 | ERNIE 5.1 | Baidu | 1 437 | 1 429–1 444 | 7 803 | Proprietary |
| 38 | Claude Opus 4.8 | Anthropic | 1 435 | 1 428–1 441 | 12 684 | Proprietary |
| 39 | Claude Opus 4.5 | Anthropic | 1 433 | 1 427–1 439 | 13 449 | Proprietary |
| 40 | Claude Sonnet 4.6 | Anthropic | 1 432 | 1 425–1 438 | 14 558 | Proprietary |
| 41 | Grok 4.5 | xAI | 1 431 | 1 424–1 439 | 7 685 | Proprietary |
| 42 | DeepSeek V4 Pro 0423 | DeepSeek | 1 429 | 1 423–1 436 | 11 822 | MIT |
| 43 | Grok 4.20 · beta-0309-reasoning | xAI | 1 428 | 1 422–1 435 | 13 551 | Proprietary |
| 44 | Grok 4.6 в режиме high | xAI | 1 428 | 1 417–1 438 | 3 939 | Proprietary |
| 45 | GLM 5 | Z.ai | 1 427 | 1 419–1 435 | 5 483 | MIT |
| 46 | Claude Sonnet 4.5 | Anthropic | 1 427 | 1 421–1 433 | 15 126 | Proprietary |
| 47 | Qwen3.6 Max Preview | Alibaba Qwen | 1 427 | 1 407–1 446 | 941 | Proprietary |
| 48 | Kimi K2.6 | Moonshot AI | 1 426 | 1 418–1 434 | 7 607 | Modified MIT |
| 49 | Claude Opus 4.5 в режиме high | Anthropic | 1 426 | 1 418–1 433 | 6 678 | Proprietary |
| 50 | GPT-5.4 | OpenAI | 1 425 | 1 419–1 432 | 13 584 | Proprietary |
| 51 | DeepSeek V4 Pro 0423 в режиме high | DeepSeek | 1 425 | 1 418–1 432 | 11 394 | MIT |
| 52 | Grok 4.20 Multi-Agent | xAI | 1 423 | 1 417–1 430 | 12 936 | Proprietary |
| 53 | Qwen3.7 Plus | Alibaba Qwen | 1 423 | 1 416–1 430 | 9 210 | Proprietary |
| 54 | Gemini 3 Flash Preview в режиме minimal | 1 421 | 1 415–1 426 | 17 940 | Proprietary | |
| 55 | DeepSeek V4 Pro 0813 в режиме high | DeepSeek | 1 419 | 1 406–1 432 | 2 255 | MIT |
| 56 | Claude Sonnet 4.5 в режиме high | Anthropic | 1 419 | 1 413–1 424 | 15 504 | Proprietary |
| 57 | Kimi K2.5 · thinking | Moonshot AI | 1 418 | 1 412–1 424 | 14 326 | Modified MIT |
| 58 | Grok 3 | xAI | 1 416 | 1 408–1 424 | 5 660 | Proprietary |
| 59 | MiMo-V2 Pro | Xiaomi | 1 416 | 1 406–1 425 | 4 693 | Proprietary |
| 60 | GPT-5.6 Terra в режиме xhigh | OpenAI | 1 415 | 1 407–1 423 | 7 049 | Proprietary |
| 61 | Grok 4.20 · beta1 | xAI | 1 415 | 1 406–1 423 | 5 125 | Proprietary |
| 62 | Gemini 3.5 Flash Lite | 1 414 | 1 406–1 422 | 6 712 | Proprietary | |
| 63 | GPT-5.1 в режиме high | OpenAI | 1 413 | 1 406–1 421 | 7 275 | Proprietary |
| 64 | ERNIE 5.0 · 0110 | Baidu | 1 412 | 1 404–1 419 | 6 586 | Proprietary |
| 65 | ChatGPT-4o (latest) | OpenAI | 1 410 | 1 404–1 415 | 14 570 | Proprietary |
| 66 | Claude Sonnet 5 в режиме high | Anthropic | 1 409 | 1 402–1 417 | 8 752 | Proprietary |
| 67 | Seed 2.0 Pro | ByteDance Seed | 1 409 | 1 403–1 415 | 15 665 | Proprietary |
| 68 | GLM 4.6 | Z.ai | 1 408 | 1 400–1 415 | 6 512 | MIT |
| 69 | Grok 4.1 · 4.1-thinking | xAI | 1 408 | 1 402–1 414 | 12 244 | Proprietary |
| 70 | GPT-5.5 · 5.5-instant | OpenAI | 1 407 | 1 398–1 416 | 5 476 | Proprietary |
| 71 | ERNIE 5.0 · preview-1203 | Baidu | 1 407 | 1 393–1 420 | 1 870 | Proprietary |
| 72 | Gemma 4 26B A4B | 1 406 | 1 388–1 424 | 1 051 | Apache 2.0 | |
| 73 | Grok 4.1 · 4.1 | xAI | 1 405 | 1 399–1 411 | 12 460 | Proprietary |
| 74 | GLM 5V Turbo | Z.ai | 1 405 | 1 392–1 418 | 2 196 | Proprietary |
| 75 | DeepSeek V3.2 Exp · exp | DeepSeek | 1 405 | 1 392–1 417 | 2 143 | MIT |
| 76 | Gemma 4 31B | 1 404 | 1 387–1 422 | 1 043 | Apache 2.0 | |
| 77 | GLM 4.5 | Z.ai | 1 404 | 1 395–1 413 | 4 291 | MIT |
| 78 | ERNIE 5.0 · preview-1022 | Baidu | 1 404 | 1 385–1 423 | 906 | Proprietary |
| 79 | Qwen3 Max · preview | Alibaba Qwen | 1 403 | 1 395–1 412 | 4 906 | Proprietary |
| 80 | GPT-5.6 Luna в режиме xhigh | OpenAI | 1 403 | 1 395–1 411 | 7 269 | Proprietary |
| 81 | Qwen3.6 Plus | Alibaba Qwen | 1 403 | 1 395–1 410 | 9 443 | Proprietary |
| 82 | MiniMax M3 | MiniMax | 1 402 | 1 395–1 408 | 11 078 | MiniMax Community License |
| 83 | R1 0528 | DeepSeek | 1 401 | 1 390–1 412 | 3 168 | MIT |
| 84 | DeepSeek V4 Flash 0423 | DeepSeek | 1 401 | 1 394–1 408 | 10 397 | MIT |
| 85 | GPT-5.2 Chat | OpenAI | 1 400 | 1 393–1 408 | 6 712 | Proprietary |
| 86 | GLM 4.7 | Z.ai | 1 400 | 1 388–1 412 | 2 363 | MIT |
| 87 | GPT-4.5 Preview | OpenAI | 1 399 | 1 388–1 411 | 2 627 | Proprietary |
| 88 | Qwen3.5 397B A17B | Alibaba Qwen | 1 399 | 1 393–1 405 | 16 670 | Apache 2.0 |
| 89 | Claude Opus 4.1 · 20250805 | Anthropic | 1 398 | 1 392–1 403 | 13 948 | Proprietary |
| 90 | Amazon Nova Experimental Chat · 26-02-10 | Amazon | 1 398 | 1 373–1 422 | 539 | Proprietary |
| 91 | MiMo-V2.5 | Xiaomi | 1 398 | 1 390–1 405 | 9 665 | MIT |
| 92 | GPT-5.1 | OpenAI | 1 396 | 1 389–1 404 | 7 920 | Proprietary |
| 93 | Claude Opus 4.1 · 20250805-thinking-16k | Anthropic | 1 396 | 1 389–1 403 | 8 931 | Proprietary |
| 94 | DeepSeek V3.2 · v3.2 | DeepSeek | 1 395 | 1 388–1 402 | 8 454 | MIT |
| 95 | LongCat-Flash-Chat · chat-2602-exp | Meituan | 1 395 | 1 386–1 403 | 5 442 | Proprietary |
| 96 | Hy3 | Tencent | 1 394 | 1 380–1 409 | 1 909 | Apache 2.0 |
| 97 | Gemini 2.5 Flash · flash | 1 394 | 1 390–1 399 | 22 600 | Proprietary | |
| 98 | Nemotron 3 Ultra | NVIDIA | 1 394 | 1 381–1 407 | 2 338 | OpenMDW-1.1 |
| 99 | MiMo-V2 Omni | Xiaomi | 1 394 | 1 384–1 404 | 4 258 | Proprietary |
| 100 | DeepSeek V3.1 · v3.1-thinking | DeepSeek | 1 393 | 1 381–1 406 | 2 198 | MIT |
| 101 | DeepSeek V4 Flash 0423 в режиме high | DeepSeek | 1 393 | 1 386–1 400 | 10 606 | MIT |
| 102 | Kimi K2.5 · instant | Moonshot AI | 1 393 | 1 378–1 408 | 1 406 | Modified MIT |
| 103 | DeepSeek V3.1 Terminus · terminus-thinking | DeepSeek | 1 392 | 1 368–1 415 | 607 | MIT |
| 104 | Grok 4 | xAI | 1 392 | 1 385–1 399 | 7 217 | Proprietary |
| 105 | Mistral Medium 3.1 | Mistral AI | 1 390 | 1 385–1 395 | 17 470 | Proprietary |
| 106 | DeepSeek V3.2 Exp · exp-thinking | DeepSeek | 1 389 | 1 375–1 404 | 1 648 | MIT |
| 107 | Hunyuan Vision 1.5 | Tencent | 1 388 | 1 358–1 417 | 367 | Proprietary |
| 108 | Inkling | Thinking Machines Lab | 1 388 | 1 379–1 396 | 6 593 | Apache 2.0 |
| 109 | DeepSeek V3.2 · v3.2-thinking | DeepSeek | 1 387 | 1 380–1 394 | 7 439 | MIT |
| 110 | Mistral Large 3 2512 | Mistral AI | 1 387 | 1 382–1 392 | 14 647 | Apache 2.0 |
| 111 | Gemini 2.5 Flash · flash-preview-09-2025 | 1 386 | 1 379–1 394 | 5 927 | Proprietary | |
| 112 | Qwen3.8 27B | Alibaba Qwen | 1 386 | 1 374–1 398 | 2 633 | Apache 2.0 |
| 113 | Gemini 3.1 Flash Lite Preview | 1 385 | 1 379–1 391 | 12 621 | Proprietary | |
| 114 | Mistral Medium 3.5 | Mistral AI | 1 383 | 1 370–1 395 | 2 443 | Modified MIT |
| 115 | Grok 4.1 Fast | xAI | 1 382 | 1 376–1 388 | 10 719 | Proprietary |
| 116 | DeepSeek V3.1 Terminus · terminus | DeepSeek | 1 381 | 1 358–1 404 | 627 | MIT |
| 117 | Grok 4 Fast · chat | xAI | 1 380 | 1 364–1 397 | 1 202 | Proprietary |
| 118 | GPT-5.2 | OpenAI | 1 379 | 1 373–1 385 | 16 393 | Proprietary |
| 119 | GPT-5 в режиме high | OpenAI | 1 379 | 1 371–1 387 | 5 790 | Proprietary |
| 120 | DeepSeek V3.1 · v3.1 | DeepSeek | 1 378 | 1 367–1 390 | 2 742 | MIT |
| 121 | Qwen3 Max · 2025-09-23 | Alibaba Qwen | 1 378 | 1 364–1 392 | 1 638 | Proprietary |
| 122 | Kimi K2 Thinking | Moonshot AI | 1 378 | 1 372–1 384 | 11 398 | Modified MIT |
| 123 | MiMo-V2 Flash · flash (non-thinking) | Xiaomi | 1 376 | 1 369–1 382 | 8 748 | MIT |
| 124 | Qwen3 235B A22B Instruct 2507 | Alibaba Qwen | 1 374 | 1 370–1 379 | 17 566 | Apache 2.0 |
| 125 | GPT-5.4 Mini в режиме high | OpenAI | 1 374 | 1 368–1 381 | 12 672 | Proprietary |
| 126 | GPT-5.2 в режиме high | OpenAI | 1 374 | 1 367–1 381 | 8 895 | Proprietary |
| 127 | o3 | OpenAI | 1 373 | 1 366–1 379 | 10 438 | Proprietary |
| 128 | Grok 4.3 | xAI | 1 371 | 1 365–1 377 | 15 759 | Proprietary |
| 129 | Claude Haiku 4.5 | Anthropic | 1 370 | 1 366–1 375 | 26 873 | Proprietary |
| 130 | Qwen3 VL 235B A22B Thinking | Alibaba Qwen | 1 370 | 1 355–1 385 | 1 434 | Apache 2.0 |
| 131 | Qwen3 235B A22B Thinking 2507 | Alibaba Qwen | 1 370 | 1 355–1 384 | 1 512 | Apache 2.0 |
| 132 | Qwen3 VL 235B A22B Instruct | Alibaba Qwen | 1 369 | 1 354–1 383 | 1 866 | Apache 2.0 |
| 133 | Claude Opus 4 · 20250514-thinking-16k | Anthropic | 1 368 | 1 360–1 376 | 6 261 | Proprietary |
| 134 | Qwen3.5-122B-A10B | Alibaba Qwen | 1 368 | 1 359–1 376 | 5 455 | Apache 2.0 |
| 135 | Grok 4 Fast · reasoning | xAI | 1 365 | 1 355–1 375 | 3 260 | Proprietary |
| 136 | GPT-4.1 | OpenAI | 1 365 | 1 358–1 372 | 8 928 | Proprietary |
| 137 | Amazon Nova Experimental Chat · 11-10 | Amazon | 1 365 | 1 356–1 374 | 4 418 | Proprietary |
| 138 | GPT-5 | OpenAI | 1 364 | 1 356–1 372 | 5 564 | Proprietary |
| 139 | MiniMax M2.1 | MiniMax | 1 362 | 1 352–1 373 | 3 151 | MIT |
| 140 | Step 3.5 Flash | StepFun | 1 361 | 1 355–1 367 | 11 335 | Apache 2.0 |
| 141 | Qwen3.5-27B | Alibaba Qwen | 1 361 | 1 352–1 369 | 5 057 | Apache 2.0 |
| 142 | LongCat-Flash-Chat · chat | Meituan | 1 360 | 1 347–1 373 | 1 938 | MIT |
| 143 | MiniMax M2.7 | MiniMax | 1 359 | 1 353–1 365 | 14 888 | Modified MIT |
| 144 | o1 · 2024-12-17 | OpenAI | 1 358 | 1 350–1 367 | 5 210 | Proprietary |
| 145 | Hy3 preview | Tencent | 1 358 | 1 341–1 375 | 1 314 | tencent-hunyuan-community |
| 146 | DeepSeek V3 0324 | DeepSeek | 1 357 | 1 350–1 363 | 7 982 | MIT |
| 147 | GPT-5.3 Chat | OpenAI | 1 356 | 1 348–1 364 | 6 390 | Proprietary |
| 148 | MiMo-V2 Flash · flash (thinking) | Xiaomi | 1 356 | 1 343–1 369 | 1 984 | MIT |
| 149 | Hunyuan T1 | Tencent | 1 354 | 1 335–1 373 | 851 | Proprietary |
| 150 | Gemini 2.5 Flash Lite · 06-17-thinking | 1 354 | 1 346–1 362 | 5 776 | Proprietary | |
| 151 | Muse Glimmer 30B | Meta | 1 352 | 1 331–1 372 | 919 | Apache-2.0 |
| 152 | Inkling Small | Thinking Machines Lab | 1 350 | 1 340–1 359 | 4 886 | Apache 2.0 |
| 153 | Qwen3 Next 80B A3B Instruct | Alibaba Qwen | 1 349 | 1 340–1 358 | 4 159 | Apache 2.0 |
| 154 | Claude Opus 4 · 20250514 | Anthropic | 1 349 | 1 342–1 357 | 7 465 | Proprietary |
| 155 | Qwen3 235B A22B · a22b-no-thinking | Alibaba Qwen | 1 349 | 1 341–1 356 | 6 586 | Apache 2.0 |
| 156 | Qwen3.5-35B-A3B | Alibaba Qwen | 1 347 | 1 339–1 356 | 5 637 | Apache 2.0 |
| 157 | Amazon Nova Experimental Chat · 26-01-10 | Amazon | 1 347 | 1 324–1 371 | 602 | Proprietary |
| 158 | Qwen3.5-Flash | Alibaba Qwen | 1 347 | 1 341–1 354 | 11 779 | Proprietary |
| 159 | R1 | DeepSeek | 1 347 | 1 337–1 356 | 3 571 | MIT |
| 160 | GLM 4.6V | Z.ai | 1 346 | 1 320–1 372 | 505 | MIT |
| 161 | Amazon Nova Experimental Chat · 12-10 | Amazon | 1 346 | 1 325–1 367 | 726 | Proprietary |
| 162 | Gemini 2.5 Flash Lite · 09-2025-no-thinking | 1 345 | 1 338–1 351 | 8 429 | Proprietary | |
| 163 | Amazon Nova Experimental Chat · 10-20 | Amazon | 1 342 | 1 329–1 355 | 2 012 | Proprietary |
| 164 | Kimi K2 0905 | Moonshot AI | 1 340 | 1 328–1 353 | 2 079 | Modified MIT |
| 165 | GPT-5 Mini в режиме high | OpenAI | 1 339 | 1 331–1 348 | 4 805 | Proprietary |
| 166 | GLM 4.5 Air | Z.ai | 1 338 | 1 330–1 346 | 5 618 | MIT |
| 167 | Nemotron 3 Super | NVIDIA | 1 336 | 1 319–1 352 | 1 183 | NVIDIA Open Model |
| 168 | Grok 3 Mini в режиме high | xAI | 1 335 | 1 324–1 346 | 2 833 | Proprietary |
| 169 | Mistral Medium 3 | Mistral AI | 1 335 | 1 327–1 343 | 5 578 | Proprietary |
| 170 | Qwen3 30B A3B Instruct 2507 | Alibaba Qwen | 1 335 | 1 325–1 344 | 4 105 | Apache 2.0 |
| 171 | Hunyuan TurboS · 20250416 | Tencent | 1 333 | 1 319–1 346 | 1 860 | Proprietary |
| 172 | Gemini 2.0 Flash | 1 333 | 1 326–1 340 | 7 728 | Proprietary | |
| 173 | Kimi K2 0711 | Moonshot AI | 1 332 | 1 323–1 341 | 4 792 | Modified MIT |
| 174 | Claude Sonnet 4 · 20250514-thinking-32k | Anthropic | 1 332 | 1 324–1 340 | 6 008 | Proprietary |
| 175 | Grok 3 Mini | xAI | 1 332 | 1 322–1 341 | 3 599 | Proprietary |
| 176 | Step 3 | StepFun | 1 331 | 1 314–1 348 | 1 082 | Apache 2.0 |
| 177 | o1 · preview | OpenAI | 1 331 | 1 321–1 341 | 4 823 | Proprietary |
| 178 | MiniMax M2.5 | MiniMax | 1 330 | 1 323–1 338 | 7 973 | Modified MIT |
| 179 | Qwen2.5 Max | Alibaba Qwen | 1 327 | 1 319–1 334 | 5 968 | Proprietary |
| 180 | GPT-5.4 Nano в режиме high | OpenAI | 1 325 | 1 318–1 331 | 12 483 | Proprietary |
| 181 | DeepSeek V3 | DeepSeek | 1 324 | 1 315–1 333 | 4 191 | DeepSeek |
| 182 | Mercury 2 | Inception Labs | 1 321 | 1 299–1 344 | 656 | Proprietary |
| 183 | Gemma 3 27B | 1 320 | 1 313–1 327 | 7 950 | Gemma | |
| 184 | Claude Sonnet 4 · 20250514 | Anthropic | 1 319 | 1 312–1 327 | 6 808 | Proprietary |
| 185 | Qwen3 Coder 480B A35B | Alibaba Qwen | 1 317 | 1 308–1 326 | 4 468 | Apache 2.0 |
| 186 | Trinity Large | Arcee AI | 1 316 | 1 307–1 324 | 5 592 | Apache 2.0 |
| 187 | Qwen3 235B A22B · a22b | Alibaba Qwen | 1 315 | 1 306–1 324 | 4 706 | Apache 2.0 |
| 188 | Claude 3.7 Sonnet · 20250219-thinking-32k | Anthropic | 1 313 | 1 306–1 321 | 6 521 | Proprietary |
| 189 | Qwen3 Next 80B A3B Thinking | Alibaba Qwen | 1 312 | 1 300–1 324 | 2 417 | Apache 2.0 |
| 190 | o4 Mini | OpenAI | 1 308 | 1 301–1 315 | 7 921 | Proprietary |
| 191 | Nova 2 Lite | Amazon | 1 308 | 1 295–1 320 | 2 197 | Proprietary |
| 192 | INTELLECT-3 | Prime Intellect | 1 308 | 1 289–1 326 | 1 065 | MIT |
| 193 | Amazon Nova Experimental Chat · 10-09 | Amazon | 1 307 | 1 283–1 331 | 483 | Proprietary |
| 194 | GLM 4.7 Flash | Z.ai | 1 307 | 1 294–1 320 | 2 013 | MIT |
| 195 | Command A | Cohere | 1 306 | 1 300–1 313 | 9 990 | CC-BY-NC-4.0 |
| 196 | Gemini 1.5 Pro · 1.5-pro-002 | 1 305 | 1 298–1 312 | 9 269 | Proprietary | |
| 197 | Mistral Small 3.2 24B | Mistral AI | 1 304 | 1 293–1 314 | 3 129 | Apache 2.0 |
| 198 | MiniMax M1 | MiniMax | 1 303 | 1 296–1 311 | 6 177 | Apache 2.0 |
| 199 | Granite 4.2 30B | IBM Granite | 1 303 | 1 279–1 326 | 735 | Apache 2.0 |
| 200 | GLM 4.5V | Z.ai | 1 303 | 1 284–1 322 | 862 | MIT |
| 201 | Solar Pro 4 | Upstage | 1 302 | 1 278–1 327 | 641 | Proprietary |
| 202 | Claude 3.7 Sonnet · 20250219 | Anthropic | 1 300 | 1 293–1 308 | 7 305 | Proprietary |
| 203 | GPT-4.1 Mini | OpenAI | 1 300 | 1 292–1 307 | 6 721 | Proprietary |
| 204 | MiniMax M2 | MiniMax | 1 300 | 1 283–1 317 | 1 172 | Apache 2.0 |
| 205 | Trinity Large Thinking | Arcee AI | 1 298 | 1 290–1 307 | 5 920 | Apache 2.0 |
| 206 | Gemma 3 12B | 1 298 | 1 276–1 319 | 697 | Gemma | |
| 207 | Gemini 2.0 Flash-Lite | 1 297 | 1 289–1 306 | 4 510 | Proprietary | |
| 208 | Nemotron 3 Nano 30B A3B | NVIDIA | 1 295 | 1 283–1 306 | 2 841 | NVIDIA Open Model |
| 209 | Step-1o Turbo | StepFun | 1 294 | 1 279–1 309 | 1 457 | Proprietary |
| 210 | Ling-flash-2.0 | inclusionAI | 1 292 | 1 275–1 309 | 1 168 | MIT |
| 211 | Step-2 16k | StepFun | 1 291 | 1 273–1 310 | 873 | Proprietary |
| 212 | gpt-oss-120b | OpenAI | 1 291 | 1 283–1 299 | 5 435 | Apache 2.0 |
| 213 | GLM-4-Plus · plus-0111 | Z.ai | 1 291 | 1 274–1 308 | 1 062 | Proprietary |
| 214 | GPT-4o (2024-05-13) | OpenAI | 1 289 | 1 283–1 296 | 17 607 | Proprietary |
| 215 | Llama 3.3 Nemotron Super 49B v1 | NVIDIA | 1 289 | 1 262–1 316 | 410 | Nvidia |
| 216 | Ring-flash-2.0 | inclusionAI | 1 286 | 1 270–1 303 | 1 220 | MIT |
| 217 | o3 Mini High | OpenAI | 1 284 | 1 273–1 294 | 3 214 | Proprietary |
| 218 | Grok 2 | xAI | 1 282 | 1 275–1 289 | 10 293 | Proprietary |
| 219 | Llama 3.3 Nemotron Super 49B v1.5 | NVIDIA | 1 281 | 1 258–1 304 | 562 | Nvidia Open |
| 220 | QwQ 32B | Alibaba Qwen | 1 279 | 1 271–1 288 | 4 434 | Apache 2.0 |
| 221 | Qwen3 32B | Alibaba Qwen | 1 279 | 1 258–1 299 | 704 | Apache 2.0 |
| 222 | Claude 3.5 Sonnet · 20241022 | Anthropic | 1 279 | 1 273–1 284 | 14 238 | Proprietary |
| 223 | Llama 3.1 Nemotron Ultra 253B v1 | NVIDIA | 1 276 | 1 253–1 300 | 532 | Nvidia Open Model |
| 224 | DeepSeek V2.5 · v2.5-1210 | DeepSeek | 1 275 | 1 259–1 291 | 1 216 | DeepSeek |
| 225 | Granite 4.2 8B | IBM Granite | 1 272 | 1 249–1 296 | 760 | Apache 2.0 |
| 226 | o3 Mini | OpenAI | 1 272 | 1 266–1 279 | 10 228 | Proprietary |
| 227 | Yi-Lightning | 01.AI | 1 272 | 1 262–1 282 | 3 980 | Proprietary |
| 228 | Gemini 1.5 Pro · advanced-0514 | 1 271 | 1 262–1 281 | 7 526 | Proprietary | |
| 229 | OLMo 3.1 32B Instruct | Ai2 | 1 271 | 1 257–1 284 | 2 013 | Apache 2.0 |
| 230 | Qwen-Plus | Alibaba Qwen | 1 270 | 1 253–1 287 | 1 091 | Proprietary |
| 231 | GPT-5 Nano в режиме high | OpenAI | 1 268 | 1 252–1 284 | 1 358 | Proprietary |
| 232 | GPT-4o (2024-08-06) | OpenAI | 1 267 | 1 259–1 276 | 7 043 | Proprietary |
| 233 | Hunyuan Turbo | Tencent | 1 267 | 1 242–1 292 | 449 | Proprietary |
| 234 | Nemotron 3.5 Lightning | NVIDIA | 1 264 | 1 249–1 279 | 2 031 | OpenMDW-1.1 |
| 235 | Gemma 3n E4B | 1 262 | 1 253–1 271 | 3 866 | Gemma | |
| 236 | Hunyuan Large | Tencent | 1 262 | 1 240–1 284 | 646 | Proprietary |
| 237 | OLMo 3 32B Think | Ai2 | 1 260 | 1 242–1 279 | 1 022 | Apache 2.0 |
| 238 | GPT-4 Turbo | OpenAI | 1 258 | 1 251–1 266 | 14 793 | Proprietary |
| 239 | Qwen3 30B A3B | Alibaba Qwen | 1 258 | 1 249–1 267 | 4 538 | Apache 2.0 |
| 240 | o1-mini | OpenAI | 1 257 | 1 250–1 264 | 8 598 | Proprietary |
| 241 | GPT-4o-mini (2024-07-18) | OpenAI | 1 253 | 1 247–1 260 | 11 283 | Proprietary |
| 242 | Hunyuan TurboS · 20250226 | Tencent | 1 253 | 1 228–1 278 | 448 | Proprietary |
| 243 | Gemini 1.5 Flash · 002 | 1 251 | 1 243–1 260 | 5 693 | Proprietary | |
| 244 | Gemini 1.5 Pro · 1.5-pro-001 | 1 251 | 1 244–1 259 | 12 133 | Proprietary | |
| 245 | Llama 3.1 405B Instruct · fp8 | Meta | 1 250 | 1 243–1 257 | 9 523 | Llama 3.1 Community |
| 246 | GLM-4-Plus · plus | Z.ai | 1 250 | 1 240–1 260 | 3 855 | Proprietary |
| 247 | Llama 4 Maverick | Meta | 1 249 | 1 242–1 257 | 6 903 | Llama 4 |
| 248 | Mistral Small 3.1 24B | Mistral AI | 1 249 | 1 241–1 257 | 5 794 | Apache 2.0 |
| 249 | Llama 3.1 Nemotron 70B Instruct | NVIDIA | 1 249 | 1 230–1 267 | 961 | Llama 3.1 |
| 250 | Granite 4.1 8B | IBM Granite | 1 248 | 1 225–1 270 | 853 | Apache 2.0 |
| 251 | Qwen2.5 Plus | Alibaba Qwen | 1 244 | 1 231–1 257 | 1 872 | Proprietary |
| 252 | GPT-4.1 Nano | OpenAI | 1 243 | 1 226–1 260 | 1 061 | Proprietary |
| 253 | GPT-4 Turbo Preview · 1106-preview | OpenAI | 1 243 | 1 235–1 250 | 15 221 | Proprietary |
| 254 | Llama 4 Scout | Meta | 1 242 | 1 234–1 251 | 5 260 | Llama |
| 255 | Magistral Medium | Mistral AI | 1 242 | 1 228–1 255 | 1 962 | Proprietary |
| 256 | Mistral Large 2407 | Mistral AI | 1 242 | 1 233–1 250 | 7 002 | Mistral Research |
| 257 | Llama 3.1 405B Instruct · bf16 | Meta | 1 241 | 1 234–1 248 | 7 230 | Llama 3.1 Community |
| 258 | Llama 3.3 70B Instruct | Meta | 1 241 | 1 235–1 247 | 9 695 | Llama-3.3 |
| 259 | Qwen Max | Alibaba Qwen | 1 240 | 1 229–1 252 | 2 396 | Qwen |
| 260 | Llama 3.1 Tulu 3 70B | Ai2 | 1 240 | 1 218–1 262 | 551 | Llama 3.1 |
| 261 | Claude 3.5 Sonnet · 20240620 | Anthropic | 1 239 | 1 231–1 246 | 13 185 | Proprietary |
| 262 | GPT-4 Turbo Preview · 0125-preview | OpenAI | 1 237 | 1 229–1 245 | 13 994 | Proprietary |
| 263 | OLMo 3.1 32B Think | Ai2 | 1 234 | 1 219–1 249 | 1 600 | Apache 2.0 |
| 264 | Grok 2 Mini | xAI | 1 233 | 1 226–1 240 | 8 614 | Proprietary |
| 265 | Gemma 3 4B | 1 233 | 1 212–1 253 | 761 | Gemma | |
| 266 | Athene 70B | Nexusflow | 1 233 | 1 222–1 243 | 3 246 | CC-BY-NC-4.0 |
| 267 | Mistral Large | Mistral AI | 1 230 | 1 222–1 239 | 5 130 | MRL |
| 268 | Granite 4.2 3B | IBM Granite | 1 230 | 1 204–1 256 | 685 | Apache 2.0 |
| 269 | DeepSeek V2.5 · v2.5 | DeepSeek | 1 230 | 1 220–1 240 | 3 664 | DeepSeek |
| 270 | Llama 3.1 70B Instruct | Meta | 1 230 | 1 222–1 237 | 8 719 | Llama 3.1 Community |
| 271 | Athene V2 Chat | Nexusflow | 1 229 | 1 220–1 238 | 4 378 | NexusFlow |
| 272 | OLMo 2 32B Instruct | Ai2 | 1 225 | 1 201–1 248 | 606 | Apache-2.0 |
| 273 | Claude 3.5 Haiku | Anthropic | 1 224 | 1 218–1 230 | 12 365 | Proprietary |
| 274 | Command R+ (08-2024) | Cohere | 1 223 | 1 209–1 238 | 1 450 | CC-BY-NC-4.0 |
| 275 | Hunyuan Standard | Tencent | 1 222 | 1 201–1 244 | 675 | Proprietary |
| 276 | Hunyuan Large Vision | Tencent | 1 221 | 1 200–1 241 | 872 | Proprietary |
| 277 | Claude 3 Opus | Anthropic | 1 220 | 1 214–1 226 | 29 359 | Proprietary |
| 278 | Jamba 1.5 Large | AI21 Labs | 1 218 | 1 203–1 233 | 1 433 | Jamba Open |
| 279 | gpt-oss-20b | OpenAI | 1 216 | 1 201–1 231 | 1 726 | Apache 2.0 |
| 280 | Gemma 2 9B IT SimPO | Princeton NLP | 1 215 | 1 201–1 229 | 1 692 | MIT |
| 281 | Granite 4.0 H Small | IBM Granite | 1 214 | 1 194–1 233 | 961 | Apache 2.0 |
| 282 | Gemma 2 27B | 1 212 | 1 206–1 219 | 12 280 | Gemma license | |
| 283 | Mercury | Inception Labs | 1 212 | 1 179–1 244 | 366 | Proprietary |
| 284 | Qwen2.5 72B Instruct | Alibaba Qwen | 1 211 | 1 203–1 219 | 6 207 | Qwen |
| 285 | Nova Pro 1.0 | Amazon | 1 210 | 1 201–1 218 | 4 518 | Proprietary |
| 286 | Nemotron-4 340B Instruct | NVIDIA | 1 208 | 1 196–1 219 | 3 252 | NVIDIA Open Model |
| 287 | Gemini 1.5 Flash · 001 | 1 208 | 1 200–1 215 | 9 767 | Proprietary | |
| 288 | Reka Core | Reka AI | 1 207 | 1 190–1 224 | 1 092 | Proprietary |
| 289 | Llama 3 70B Instruct | Meta | 1 195 | 1 188–1 203 | 23 202 | Llama 3 Community |
| 290 | Llama 3.1 Nemotron 51B Instruct | NVIDIA | 1 191 | 1 167–1 215 | 551 | Llama 3.1 |
| 291 | Command R+ | Cohere | 1 190 | 1 181–1 198 | 11 264 | CC-BY-NC-4.0 |
| 292 | GLM-4 | Z.ai | 1 185 | 1 171–1 200 | 1 605 | Proprietary |
| 293 | Gemini 1.5 Flash-8B | 1 185 | 1 177–1 193 | 5 700 | Proprietary | |
| 294 | Gemma 2 9B | 1 185 | 1 178–1 192 | 9 128 | Gemma license | |
| 295 | GPT-4 · 0314 | OpenAI | 1 184 | 1 175–1 194 | 8 080 | Proprietary |
| 296 | Mistral Small 3 | Mistral AI | 1 184 | 1 172–1 195 | 2 680 | Apache 2.0 |
| 297 | Aya Expanse 32B | Cohere | 1 183 | 1 174–1 193 | 4 344 | CC-BY-NC-4.0 |
| 298 | GPT-4 · 0613 | OpenAI | 1 180 | 1 172–1 188 | 13 910 | Proprietary |
| 299 | Reka Flash (2024-09) | Reka AI | 1 177 | 1 161–1 194 | 1 115 | Proprietary |
| 300 | Nova Lite 1.0 | Amazon | 1 176 | 1 166–1 186 | 3 418 | Proprietary |
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.