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 «не английский», 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 | 1 505 | 1 494–1 515 | 3 374 | Proprietary |
| 2 | Claude Opus 5 в режиме high | Anthropic | 1 501 | 1 496–1 506 | 24 947 | Proprietary |
| 3 | Claude Opus 5 в режиме max | Anthropic | 1 499 | 1 492–1 505 | 12 121 | Proprietary |
| 4 | Claude Opus 4.6 в режиме high | Anthropic | 1 490 | 1 485–1 494 | 39 972 | Proprietary |
| 5 | Muse Spark 1.3 в режиме max | Meta | 1 489 | 1 478–1 501 | 2 816 | Proprietary |
| 6 | Claude Opus 4.6 | Anthropic | 1 486 | 1 482–1 490 | 41 581 | Proprietary |
| 7 | Gemini 3.7 Flash в режиме high | 1 486 | 1 475–1 497 | 3 275 | Proprietary | |
| 8 | Claude Fable 5 | Anthropic | 1 484 | 1 479–1 490 | 16 958 | Proprietary |
| 9 | Muse Spark 1.2 в режиме xhigh | Meta | 1 483 | 1 469–1 497 | 1 839 | Proprietary |
| 10 | Gemini 3.8 Flash в режиме high | 1 483 | 1 472–1 493 | 3 090 | Proprietary | |
| 11 | Claude Opus 4.7 в режиме high | Anthropic | 1 481 | 1 477–1 486 | 31 977 | Proprietary |
| 12 | Gemini 3.1 Pro Preview | 1 477 | 1 473–1 481 | 59 109 | Proprietary | |
| 13 | Gemini 3.5 Flash в режиме high | 1 477 | 1 471–1 482 | 21 659 | Proprietary | |
| 14 | Claude Opus 4.7 | Anthropic | 1 475 | 1 470–1 479 | 32 815 | Proprietary |
| 15 | Gemini 3 Pro | 1 474 | 1 469–1 479 | 24 208 | Proprietary | |
| 16 | Muse Spark 1.1 | Meta | 1 473 | 1 468–1 479 | 16 196 | Proprietary |
| 17 | Qwen3.7 Max | Alibaba Qwen | 1 472 | 1 458–1 485 | 1 897 | Proprietary |
| 18 | Qwen3.8 Max (0902) | Alibaba Qwen | 1 470 | 1 463–1 477 | 10 082 | Proprietary |
| 19 | Gemini 3.5 Flash в режиме medium | 1 469 | 1 463–1 474 | 20 781 | Proprietary | |
| 20 | Gemini 3.6 Flash в режиме high | 1 469 | 1 463–1 475 | 15 134 | Proprietary | |
| 21 | GPT-5.5 в режиме high | OpenAI | 1 467 | 1 462–1 472 | 35 398 | Proprietary |
| 22 | Qwen3.5 Max | Alibaba Qwen | 1 466 | 1 460–1 472 | 11 738 | Proprietary |
| 23 | GPT-5.4 в режиме high | OpenAI | 1 465 | 1 460–1 470 | 33 029 | Proprietary |
| 24 | Muse Spark | Meta | 1 463 | 1 456–1 471 | 7 097 | Proprietary |
| 25 | GLM 5.3 Flash | Z.ai | 1 461 | 1 453–1 469 | 6 147 | MIT |
| 26 | Kimi K3 в режиме max | Moonshot AI | 1 461 | 1 455–1 467 | 12 798 | Kimi K3 license |
| 27 | Gemini 3 Flash Preview | 1 459 | 1 454–1 465 | 18 112 | Proprietary | |
| 28 | GLM 5.3 в режиме max | Z.ai | 1 459 | 1 451–1 467 | 6 475 | MIT |
| 29 | GPT-5.5 · 5.5 | OpenAI | 1 459 | 1 454–1 463 | 36 079 | Proprietary |
| 30 | GLM 5.2 в режиме max | Z.ai | 1 455 | 1 450–1 461 | 21 161 | MIT |
| 31 | ERNIE 5.1 | Baidu | 1 455 | 1 449–1 461 | 20 073 | Proprietary |
| 32 | Claude Opus 4.8 в режиме high | Anthropic | 1 452 | 1 447–1 457 | 29 293 | Proprietary |
| 33 | Gemini 2.5 Pro | 1 451 | 1 448–1 455 | 69 533 | Proprietary | |
| 34 | GPT-5.6 Sol в режиме xhigh | OpenAI | 1 450 | 1 444–1 455 | 15 900 | Proprietary |
| 35 | GPT-5.4 | OpenAI | 1 449 | 1 444–1 454 | 34 501 | Proprietary |
| 36 | GLM 5.1 | Z.ai | 1 448 | 1 444–1 453 | 27 294 | MIT |
| 37 | MiMo-V2.5-Pro | Xiaomi | 1 448 | 1 443–1 452 | 34 753 | MIT |
| 38 | Kimi K2.6 | Moonshot AI | 1 446 | 1 441–1 452 | 20 342 | Modified MIT |
| 39 | Qwen3.7 Plus | Alibaba Qwen | 1 444 | 1 439–1 450 | 22 671 | Proprietary |
| 40 | Claude Opus 4.8 | Anthropic | 1 443 | 1 438–1 448 | 29 507 | Proprietary |
| 41 | Grok 4.5 | xAI | 1 442 | 1 436–1 447 | 17 689 | Proprietary |
| 42 | Seed 2.0 Pro | ByteDance Seed | 1 441 | 1 437–1 445 | 40 894 | Proprietary |
| 43 | Grok 4.20 · beta-0309-reasoning | xAI | 1 441 | 1 436–1 446 | 33 537 | Proprietary |
| 44 | Claude Sonnet 4.6 | Anthropic | 1 440 | 1 436–1 445 | 36 233 | Proprietary |
| 45 | GPT-5.6 Terra в режиме xhigh | OpenAI | 1 440 | 1 434–1 446 | 16 498 | Proprietary |
| 46 | Grok 4.20 Multi-Agent | xAI | 1 440 | 1 435–1 444 | 32 964 | Proprietary |
| 47 | DeepSeek V4 Pro 0423 | DeepSeek | 1 439 | 1 434–1 444 | 30 240 | MIT |
| 48 | Gemini 3 Flash Preview в режиме minimal | 1 437 | 1 434–1 441 | 47 620 | Proprietary | |
| 49 | Claude Opus 4.5 | Anthropic | 1 437 | 1 433–1 441 | 39 814 | Proprietary |
| 50 | Qwen3.6 Max Preview | Alibaba Qwen | 1 437 | 1 425–1 448 | 2 801 | Proprietary |
| 51 | ERNIE 5.0 · 0110 | Baidu | 1 436 | 1 431–1 441 | 20 294 | Proprietary |
| 52 | ERNIE 5.0 · preview-1203 | Baidu | 1 435 | 1 426–1 443 | 5 568 | Proprietary |
| 53 | Amazon Nova Experimental Chat · 26-02-10 | Amazon | 1 434 | 1 421–1 446 | 2 102 | Proprietary |
| 54 | Claude Opus 4.5 в режиме high | Anthropic | 1 433 | 1 428–1 438 | 21 774 | Proprietary |
| 55 | Kimi K2.5 · thinking | Moonshot AI | 1 433 | 1 429–1 437 | 40 413 | Modified MIT |
| 56 | DeepSeek V4 Pro 0423 в режиме high | DeepSeek | 1 432 | 1 427–1 437 | 28 809 | MIT |
| 57 | Qwen3.5 397B A17B | Alibaba Qwen | 1 432 | 1 428–1 436 | 42 333 | Apache 2.0 |
| 58 | Grok 4.20 · beta1 | xAI | 1 432 | 1 426–1 437 | 14 327 | Proprietary |
| 59 | DeepSeek V4 Pro 0813 в режиме high | DeepSeek | 1 431 | 1 423–1 439 | 5 499 | MIT |
| 60 | GPT-5.1 в режиме high | OpenAI | 1 430 | 1 426–1 435 | 24 027 | Proprietary |
| 61 | Inkling | Thinking Machines Lab | 1 430 | 1 424–1 436 | 15 116 | Apache 2.0 |
| 62 | GLM 5 | Z.ai | 1 430 | 1 424–1 435 | 15 300 | MIT |
| 63 | Qwen3 Max · preview | Alibaba Qwen | 1 430 | 1 424–1 435 | 15 710 | Proprietary |
| 64 | Claude Sonnet 5 в режиме high | Anthropic | 1 429 | 1 424–1 435 | 20 180 | Proprietary |
| 65 | Qwen3.8 27B | Alibaba Qwen | 1 429 | 1 422–1 437 | 6 518 | Apache 2.0 |
| 66 | Gemma 4 31B | 1 429 | 1 420–1 439 | 3 622 | Apache 2.0 | |
| 67 | Gemini 3.5 Flash Lite | 1 428 | 1 422–1 434 | 15 302 | Proprietary | |
| 68 | GPT-6 Astra в режиме max | OpenAI | 1 428 | 1 412–1 443 | 1 554 | Proprietary |
| 69 | ERNIE 5.0 · preview-1022 | Baidu | 1 427 | 1 416–1 439 | 2 504 | Proprietary |
| 70 | Nemotron 3 Ultra | NVIDIA | 1 427 | 1 418–1 435 | 5 736 | OpenMDW-1.1 |
| 71 | GLM 4.6 | Z.ai | 1 427 | 1 422–1 432 | 20 015 | MIT |
| 72 | GPT-5.2 Chat | OpenAI | 1 426 | 1 421–1 431 | 18 951 | Proprietary |
| 73 | Grok 4.1 · 4.1-thinking | xAI | 1 426 | 1 422–1 430 | 36 797 | Proprietary |
| 74 | Hy3 | Tencent | 1 426 | 1 417–1 435 | 4 926 | Apache 2.0 |
| 75 | Claude Sonnet 4.5 | Anthropic | 1 425 | 1 421–1 428 | 44 960 | Proprietary |
| 76 | Qwen3.6 Plus | Alibaba Qwen | 1 424 | 1 419–1 429 | 24 818 | Proprietary |
| 77 | Grok 4.6 в режиме high | xAI | 1 422 | 1 415–1 429 | 9 102 | Proprietary |
| 78 | Gemma 4 26B A4B | 1 422 | 1 412–1 431 | 3 638 | Apache 2.0 | |
| 79 | Grok 4.1 · 4.1 | xAI | 1 421 | 1 418–1 425 | 38 133 | Proprietary |
| 80 | MiniMax M3 | MiniMax | 1 421 | 1 416–1 426 | 27 916 | MiniMax Community License |
| 81 | GLM 5V Turbo | Z.ai | 1 420 | 1 412–1 429 | 5 369 | Proprietary |
| 82 | DeepSeek V4 Flash 0423 | DeepSeek | 1 419 | 1 414–1 424 | 26 993 | MIT |
| 83 | ChatGPT-4o (latest) | OpenAI | 1 419 | 1 416–1 423 | 43 832 | Proprietary |
| 84 | GLM 4.7 | Z.ai | 1 418 | 1 410–1 425 | 7 006 | MIT |
| 85 | GLM 4.5 | Z.ai | 1 417 | 1 411–1 423 | 13 631 | MIT |
| 86 | GPT-5.5 · 5.5-instant | OpenAI | 1 417 | 1 411–1 423 | 13 489 | Proprietary |
| 87 | Claude Sonnet 4.5 в режиме high | Anthropic | 1 417 | 1 413–1 420 | 45 869 | Proprietary |
| 88 | MiMo-V2 Pro | Xiaomi | 1 417 | 1 411–1 422 | 13 757 | Proprietary |
| 89 | GPT-5.6 Luna в режиме xhigh | OpenAI | 1 416 | 1 410–1 422 | 16 737 | Proprietary |
| 90 | GPT-5.1 | OpenAI | 1 413 | 1 409–1 418 | 25 591 | Proprietary |
| 91 | GPT-4.5 Preview | OpenAI | 1 413 | 1 405–1 421 | 6 122 | Proprietary |
| 92 | Mistral Large 3 2512 | Mistral AI | 1 412 | 1 409–1 416 | 39 976 | Apache 2.0 |
| 93 | R1 0528 | DeepSeek | 1 412 | 1 405–1 418 | 9 954 | MIT |
| 94 | Grok 3 | xAI | 1 410 | 1 405–1 416 | 15 350 | Proprietary |
| 95 | Gemini 3.1 Flash Lite Preview | 1 410 | 1 406–1 415 | 32 534 | Proprietary | |
| 96 | Qwen3 235B A22B Instruct 2507 | Alibaba Qwen | 1 410 | 1 407–1 413 | 55 480 | Apache 2.0 |
| 97 | Gemini 2.5 Flash · flash | 1 410 | 1 407–1 413 | 69 018 | Proprietary | |
| 98 | DeepSeek V3.2 · v3.2 | DeepSeek | 1 409 | 1 405–1 414 | 27 427 | MIT |
| 99 | Mistral Medium 3.1 | Mistral AI | 1 409 | 1 405–1 412 | 52 778 | Proprietary |
| 100 | DeepSeek V4 Flash 0423 в режиме high | DeepSeek | 1 409 | 1 404–1 414 | 26 897 | MIT |
| 101 | Qwen3 Next 80B A3B Instruct | Alibaba Qwen | 1 407 | 1 401–1 413 | 12 864 | Apache 2.0 |
| 102 | DeepSeek V3.1 Terminus · terminus | DeepSeek | 1 407 | 1 394–1 420 | 2 029 | MIT |
| 103 | DeepSeek V3.1 Terminus · terminus-thinking | DeepSeek | 1 406 | 1 393–1 419 | 1 913 | MIT |
| 104 | GPT-5.2 | OpenAI | 1 406 | 1 402–1 410 | 43 368 | Proprietary |
| 105 | Qwen3 VL 235B A22B Instruct | Alibaba Qwen | 1 406 | 1 397–1 414 | 6 339 | Apache 2.0 |
| 106 | GPT-5.4 Mini в режиме high | OpenAI | 1 406 | 1 401–1 410 | 31 940 | Proprietary |
| 107 | Kimi K2.5 · instant | Moonshot AI | 1 406 | 1 397–1 414 | 4 864 | Modified MIT |
| 108 | GPT-5.2 в режиме high | OpenAI | 1 405 | 1 401–1 410 | 27 090 | Proprietary |
| 109 | Amazon Nova Experimental Chat · 11-10 | Amazon | 1 405 | 1 400–1 411 | 14 872 | Proprietary |
| 110 | DeepSeek V3.2 Exp · exp | DeepSeek | 1 405 | 1 397–1 413 | 6 787 | MIT |
| 111 | Mistral Medium 3.5 | Mistral AI | 1 405 | 1 397–1 413 | 5 706 | Modified MIT |
| 112 | Claude Opus 4.1 · 20250805 | Anthropic | 1 405 | 1 401–1 408 | 44 216 | Proprietary |
| 113 | DeepSeek V3.2 Exp · exp-thinking | DeepSeek | 1 404 | 1 396–1 413 | 5 061 | MIT |
| 114 | MiMo-V2.5 | Xiaomi | 1 404 | 1 399–1 409 | 24 288 | MIT |
| 115 | LongCat-Flash-Chat · chat | Meituan | 1 404 | 1 396–1 412 | 6 440 | MIT |
| 116 | Claude Opus 4.1 · 20250805-thinking-16k | Anthropic | 1 404 | 1 400–1 408 | 28 186 | Proprietary |
| 117 | MiMo-V2 Omni | Xiaomi | 1 403 | 1 396–1 410 | 10 406 | Proprietary |
| 118 | Grok 4 | xAI | 1 402 | 1 398–1 407 | 21 868 | Proprietary |
| 119 | o3 | OpenAI | 1 402 | 1 397–1 406 | 31 411 | Proprietary |
| 120 | Qwen3 Max · 2025-09-23 | Alibaba Qwen | 1 400 | 1 392–1 409 | 5 070 | Proprietary |
| 121 | Qwen3.5-122B-A10B | Alibaba Qwen | 1 400 | 1 395–1 406 | 15 712 | Apache 2.0 |
| 122 | DeepSeek V3.1 · v3.1-thinking | DeepSeek | 1 400 | 1 392–1 408 | 6 799 | MIT |
| 123 | DeepSeek V3.1 · v3.1 | DeepSeek | 1 400 | 1 392–1 407 | 8 614 | MIT |
| 124 | Amazon Nova Experimental Chat · 12-10 | Amazon | 1 399 | 1 387–1 411 | 2 166 | Proprietary |
| 125 | DeepSeek V3.2 · v3.2-thinking | DeepSeek | 1 398 | 1 394–1 403 | 23 859 | MIT |
| 126 | LongCat-Flash-Chat · chat-2602-exp | Meituan | 1 398 | 1 392–1 404 | 14 804 | Proprietary |
| 127 | Kimi K2 Thinking | Moonshot AI | 1 398 | 1 394–1 401 | 35 617 | Modified MIT |
| 128 | Qwen3 235B A22B Thinking 2507 | Alibaba Qwen | 1 397 | 1 389–1 406 | 4 899 | Apache 2.0 |
| 129 | Inkling Small | Thinking Machines Lab | 1 397 | 1 391–1 404 | 10 988 | Apache 2.0 |
| 130 | GPT-5 в режиме high | OpenAI | 1 397 | 1 391–1 402 | 17 931 | Proprietary |
| 131 | Grok 4 Fast · chat | xAI | 1 396 | 1 387–1 406 | 3 691 | Proprietary |
| 132 | Gemini 2.5 Flash · flash-preview-09-2025 | 1 396 | 1 391–1 401 | 18 534 | Proprietary | |
| 133 | Hunyuan T1 | Tencent | 1 395 | 1 383–1 406 | 2 636 | Proprietary |
| 134 | GPT-5 | OpenAI | 1 395 | 1 389–1 400 | 17 747 | Proprietary |
| 135 | Hy3 preview | Tencent | 1 394 | 1 384–1 404 | 3 392 | tencent-hunyuan-community |
| 136 | MiMo-V2 Flash · flash (non-thinking) | Xiaomi | 1 391 | 1 387–1 396 | 26 916 | MIT |
| 137 | Grok 4.1 Fast | xAI | 1 391 | 1 387–1 395 | 31 757 | Proprietary |
| 138 | Qwen3.5-27B | Alibaba Qwen | 1 389 | 1 383–1 394 | 15 002 | Apache 2.0 |
| 139 | Muse Glimmer 30B | Meta | 1 388 | 1 376–1 401 | 2 160 | Apache-2.0 |
| 140 | Grok 4 Fast · reasoning | xAI | 1 388 | 1 382–1 394 | 9 967 | Proprietary |
| 141 | Grok 4.3 | xAI | 1 387 | 1 383–1 392 | 36 025 | Proprietary |
| 142 | Step 3.5 Flash | StepFun | 1 386 | 1 382–1 390 | 33 007 | Apache 2.0 |
| 143 | Qwen3.5-Flash | Alibaba Qwen | 1 385 | 1 381–1 390 | 32 942 | Proprietary |
| 144 | Qwen3 235B A22B · a22b-no-thinking | Alibaba Qwen | 1 384 | 1 379–1 390 | 20 828 | Apache 2.0 |
| 145 | Amazon Nova Experimental Chat · 26-01-10 | Amazon | 1 384 | 1 371–1 396 | 1 997 | Proprietary |
| 146 | Hunyuan Vision 1.5 | Tencent | 1 382 | 1 366–1 399 | 1 216 | Proprietary |
| 147 | MiniMax M2.7 | MiniMax | 1 382 | 1 378–1 386 | 39 775 | Modified MIT |
| 148 | GPT-5.3 Chat | OpenAI | 1 381 | 1 376–1 387 | 18 163 | Proprietary |
| 149 | Qwen3.5-35B-A3B | Alibaba Qwen | 1 380 | 1 374–1 385 | 16 000 | Apache 2.0 |
| 150 | Qwen3 VL 235B A22B Thinking | Alibaba Qwen | 1 377 | 1 369–1 386 | 4 325 | Apache 2.0 |
| 151 | MiniMax M2.1 | MiniMax | 1 377 | 1 370–1 383 | 10 116 | MIT |
| 152 | Claude Haiku 4.5 | Anthropic | 1 376 | 1 373–1 380 | 72 573 | Proprietary |
| 153 | MiMo-V2 Flash · flash (thinking) | Xiaomi | 1 373 | 1 365–1 380 | 6 514 | MIT |
| 154 | Amazon Nova Experimental Chat · 10-20 | Amazon | 1 372 | 1 365–1 380 | 6 434 | Proprietary |
| 155 | Kimi K2 0905 | Moonshot AI | 1 372 | 1 364–1 379 | 6 804 | Modified MIT |
| 156 | Qwen3 30B A3B Instruct 2507 | Alibaba Qwen | 1 372 | 1 366–1 377 | 13 499 | Apache 2.0 |
| 157 | GPT-4.1 | OpenAI | 1 370 | 1 365–1 374 | 26 739 | Proprietary |
| 158 | Gemini 2.5 Flash Lite · 09-2025-no-thinking | 1 369 | 1 364–1 373 | 26 732 | Proprietary | |
| 159 | GLM 4.5 Air | Z.ai | 1 366 | 1 361–1 371 | 17 243 | MIT |
| 160 | GPT-5 Mini в режиме high | OpenAI | 1 363 | 1 357–1 368 | 15 175 | Proprietary |
| 161 | Hunyuan TurboS · 20250416 | Tencent | 1 362 | 1 354–1 370 | 5 193 | Proprietary |
| 162 | Claude Opus 4 · 20250514-thinking-16k | Anthropic | 1 362 | 1 356–1 367 | 20 329 | Proprietary |
| 163 | GPT-5.4 Nano в режиме high | OpenAI | 1 360 | 1 355–1 365 | 31 195 | Proprietary |
| 164 | Gemini 2.5 Flash Lite · 06-17-thinking | 1 360 | 1 355–1 365 | 18 225 | Proprietary | |
| 165 | GLM 4.6V | Z.ai | 1 360 | 1 345–1 374 | 1 620 | MIT |
| 166 | o1 · 2024-12-17 | OpenAI | 1 358 | 1 352–1 364 | 11 491 | Proprietary |
| 167 | DeepSeek V3 0324 | DeepSeek | 1 358 | 1 353–1 363 | 23 442 | MIT |
| 168 | Kimi K2 0711 | Moonshot AI | 1 358 | 1 352–1 363 | 15 366 | Modified MIT |
| 169 | R1 | DeepSeek | 1 356 | 1 350–1 363 | 7 803 | MIT |
| 170 | Nemotron 3 Super | NVIDIA | 1 356 | 1 347–1 365 | 4 427 | NVIDIA Open Model |
| 171 | Mistral Medium 3 | Mistral AI | 1 352 | 1 347–1 358 | 17 577 | Proprietary |
| 172 | Qwen2.5 Max | Alibaba Qwen | 1 352 | 1 347–1 358 | 14 251 | Proprietary |
| 173 | Grok 3 Mini в режиме high | xAI | 1 352 | 1 345–1 358 | 8 438 | Proprietary |
| 174 | Claude Opus 4 · 20250514 | Anthropic | 1 351 | 1 346–1 357 | 23 893 | Proprietary |
| 175 | Solar Pro 4 | Upstage | 1 351 | 1 335–1 367 | 1 440 | Proprietary |
| 176 | gpt-oss-120b | OpenAI | 1 351 | 1 345–1 356 | 17 001 | Apache 2.0 |
| 177 | Qwen3 235B A22B · a22b | Alibaba Qwen | 1 349 | 1 343–1 354 | 13 511 | Apache 2.0 |
| 178 | Grok 3 Mini | xAI | 1 348 | 1 342–1 355 | 11 020 | Proprietary |
| 179 | Qwen3 Coder 480B A35B | Alibaba Qwen | 1 345 | 1 339–1 351 | 14 157 | Apache 2.0 |
| 180 | Gemini 2.0 Flash | 1 342 | 1 337–1 347 | 19 596 | Proprietary | |
| 181 | Qwen3 Next 80B A3B Thinking | Alibaba Qwen | 1 342 | 1 335–1 349 | 7 525 | Apache 2.0 |
| 182 | Granite 4.2 30B | IBM Granite | 1 341 | 1 327–1 355 | 1 857 | Apache 2.0 |
| 183 | Amazon Nova Experimental Chat · 10-09 | Amazon | 1 340 | 1 326–1 355 | 1 535 | Proprietary |
| 184 | Nova 2 Lite | Amazon | 1 338 | 1 330–1 346 | 7 056 | Proprietary |
| 185 | MiniMax M2.5 | MiniMax | 1 337 | 1 332–1 342 | 22 777 | Modified MIT |
| 186 | o4 Mini | OpenAI | 1 336 | 1 331–1 341 | 23 891 | Proprietary |
| 187 | Ling-flash-2.0 | inclusionAI | 1 335 | 1 326–1 345 | 3 853 | MIT |
| 188 | Gemma 3 27B | 1 334 | 1 330–1 339 | 23 473 | Gemma | |
| 189 | GLM 4.7 Flash | Z.ai | 1 333 | 1 325–1 340 | 7 032 | MIT |
| 190 | Mercury 2 | Inception Labs | 1 332 | 1 319–1 346 | 1 872 | Proprietary |
| 191 | Claude Sonnet 4 · 20250514-thinking-32k | Anthropic | 1 332 | 1 327–1 337 | 19 194 | Proprietary |
| 192 | Claude Sonnet 4 · 20250514 | Anthropic | 1 330 | 1 325–1 335 | 21 848 | Proprietary |
| 193 | INTELLECT-3 | Prime Intellect | 1 329 | 1 318–1 340 | 2 938 | MIT |
| 194 | Trinity Large Thinking | Arcee AI | 1 328 | 1 322–1 334 | 15 198 | Apache 2.0 |
| 195 | Step 3 | StepFun | 1 327 | 1 317–1 337 | 3 503 | Apache 2.0 |
| 196 | Gemini 2.0 Flash-Lite | 1 323 | 1 316–1 329 | 10 064 | Proprietary | |
| 197 | Nemotron 3 Nano 30B A3B | NVIDIA | 1 320 | 1 313–1 327 | 9 019 | NVIDIA Open Model |
| 198 | MiniMax M1 | MiniMax | 1 320 | 1 315–1 325 | 19 202 | Apache 2.0 |
| 199 | o3 Mini High | OpenAI | 1 319 | 1 311–1 326 | 7 459 | Proprietary |
| 200 | Trinity Large | Arcee AI | 1 318 | 1 312–1 323 | 16 378 | Apache 2.0 |
| 201 | Gemma 3 12B | 1 318 | 1 304–1 332 | 1 499 | Gemma | |
| 202 | GPT-4.1 Mini | OpenAI | 1 318 | 1 312–1 323 | 20 604 | Proprietary |
| 203 | Qwen3 32B | Alibaba Qwen | 1 317 | 1 303–1 330 | 1 696 | Apache 2.0 |
| 204 | Llama 3.3 Nemotron Super 49B v1.5 | NVIDIA | 1 316 | 1 303–1 329 | 1 934 | Nvidia Open |
| 205 | MiniMax M2 | MiniMax | 1 316 | 1 306–1 326 | 3 773 | Apache 2.0 |
| 206 | DeepSeek V3 | DeepSeek | 1 315 | 1 309–1 322 | 8 904 | DeepSeek |
| 207 | Mistral Small 3.2 24B | Mistral AI | 1 315 | 1 309–1 322 | 9 863 | Apache 2.0 |
| 208 | o1 · preview | OpenAI | 1 315 | 1 308–1 321 | 15 231 | Proprietary |
| 209 | GLM-4-Plus · plus-0111 | Z.ai | 1 314 | 1 302–1 326 | 2 294 | Proprietary |
| 210 | Command A | Cohere | 1 313 | 1 308–1 317 | 29 304 | CC-BY-NC-4.0 |
| 211 | Step-1o Turbo | StepFun | 1 312 | 1 304–1 321 | 4 890 | Proprietary |
| 212 | GPT-5 Nano в режиме high | OpenAI | 1 312 | 1 304–1 321 | 4 551 | Proprietary |
| 213 | Gemini 1.5 Pro · 1.5-pro-002 | 1 312 | 1 307–1 317 | 25 473 | Proprietary | |
| 214 | Qwen-Plus | Alibaba Qwen | 1 310 | 1 298–1 321 | 2 394 | Proprietary |
| 215 | QwQ 32B | Alibaba Qwen | 1 305 | 1 300–1 311 | 12 198 | Apache 2.0 |
| 216 | GLM 4.5V | Z.ai | 1 303 | 1 292–1 314 | 2 790 | MIT |
| 217 | o3 Mini | OpenAI | 1 300 | 1 295–1 304 | 28 044 | Proprietary |
| 218 | Nemotron 3.5 Lightning | NVIDIA | 1 298 | 1 289–1 308 | 4 865 | OpenMDW-1.1 |
| 219 | Ring-flash-2.0 | inclusionAI | 1 298 | 1 288–1 307 | 3 885 | MIT |
| 220 | Hunyuan TurboS · 20250226 | Tencent | 1 298 | 1 280–1 316 | 910 | Proprietary |
| 221 | Granite 4.2 8B | IBM Granite | 1 297 | 1 282–1 312 | 1 742 | Apache 2.0 |
| 222 | Claude 3.7 Sonnet · 20250219-thinking-32k | Anthropic | 1 296 | 1 291–1 301 | 19 205 | Proprietary |
| 223 | Qwen3 30B A3B | Alibaba Qwen | 1 295 | 1 289–1 301 | 13 800 | Apache 2.0 |
| 224 | Step-2 16k | StepFun | 1 290 | 1 277–1 302 | 2 052 | Proprietary |
| 225 | o1-mini | OpenAI | 1 288 | 1 284–1 293 | 24 629 | Proprietary |
| 226 | Claude 3.7 Sonnet · 20250219 | Anthropic | 1 286 | 1 281–1 291 | 20 402 | Proprietary |
| 227 | Hunyuan Turbo | Tencent | 1 285 | 1 268–1 302 | 971 | Proprietary |
| 228 | Gemma 3n E4B | 1 285 | 1 278–1 291 | 12 011 | Gemma | |
| 229 | Molmo 2 8B | Ai2 | 1 284 | 1 258–1 311 | 466 | Apache 2.0 |
| 230 | Claude 3.5 Sonnet · 20241022 | Anthropic | 1 283 | 1 279–1 287 | 42 653 | Proprietary |
| 231 | GPT-4o (2024-05-13) | OpenAI | 1 283 | 1 278–1 287 | 52 655 | Proprietary |
| 232 | Llama 3.1 Nemotron Ultra 253B v1 | NVIDIA | 1 282 | 1 265–1 299 | 1 058 | Nvidia Open Model |
| 233 | Grok 2 | xAI | 1 282 | 1 277–1 287 | 29 136 | Proprietary |
| 234 | Gemini 1.5 Flash · 002 | 1 278 | 1 272–1 284 | 16 538 | Proprietary | |
| 235 | OLMo 3.1 32B Instruct | Ai2 | 1 276 | 1 268–1 284 | 6 987 | Apache 2.0 |
| 236 | GLM-4-Plus · plus | Z.ai | 1 275 | 1 269–1 282 | 12 935 | Proprietary |
| 237 | Gemini 1.5 Pro · advanced-0514 | 1 274 | 1 267–1 280 | 23 830 | Proprietary | |
| 238 | DeepSeek V2.5 · v2.5-1210 | DeepSeek | 1 273 | 1 262–1 284 | 3 019 | DeepSeek |
| 239 | Gemma 3 4B | 1 273 | 1 259–1 287 | 1 599 | Gemma | |
| 240 | Granite 4.2 3B | IBM Granite | 1 272 | 1 256–1 288 | 1 777 | Apache 2.0 |
| 241 | Claude 3.5 Sonnet · 20240620 | Anthropic | 1 270 | 1 265–1 275 | 37 986 | Proprietary |
| 242 | Llama 4 Maverick | Meta | 1 269 | 1 264–1 274 | 20 487 | Llama 4 |
| 243 | Yi-Lightning | 01.AI | 1 269 | 1 262–1 275 | 13 682 | Proprietary |
| 244 | gpt-oss-20b | OpenAI | 1 268 | 1 260–1 276 | 5 879 | Apache 2.0 |
| 245 | Athene V2 Chat | Nexusflow | 1 268 | 1 262–1 274 | 11 063 | NexusFlow |
| 246 | GPT-4o-mini (2024-07-18) | OpenAI | 1 266 | 1 262–1 271 | 30 968 | Proprietary |
| 247 | Hunyuan Large | Tencent | 1 265 | 1 250–1 280 | 1 483 | Proprietary |
| 248 | Qwen2.5 Plus | Alibaba Qwen | 1 265 | 1 256–1 274 | 4 383 | Proprietary |
| 249 | Gemini 1.5 Pro · 1.5-pro-001 | 1 264 | 1 259–1 270 | 36 600 | Proprietary | |
| 250 | Mercury | Inception Labs | 1 263 | 1 245–1 281 | 1 089 | Proprietary |
| 251 | Qwen Max | Alibaba Qwen | 1 263 | 1 255–1 270 | 8 131 | Qwen |
| 252 | Granite 4.1 8B | IBM Granite | 1 262 | 1 248–1 276 | 2 072 | Apache 2.0 |
| 253 | GPT-4o (2024-08-06) | OpenAI | 1 262 | 1 256–1 267 | 20 213 | Proprietary |
| 254 | Hunyuan Standard | Tencent | 1 261 | 1 247–1 275 | 1 592 | Proprietary |
| 255 | GPT-4.1 Nano | OpenAI | 1 260 | 1 249–1 271 | 2 575 | Proprietary |
| 256 | Claude 3 Opus | Anthropic | 1 258 | 1 253–1 262 | 86 483 | Proprietary |
| 257 | Grok 2 Mini | xAI | 1 257 | 1 252–1 262 | 24 138 | Proprietary |
| 258 | OLMo 3 32B Think | Ai2 | 1 256 | 1 245–1 267 | 3 281 | Apache 2.0 |
| 259 | Mistral Small 3.1 24B | Mistral AI | 1 255 | 1 249–1 260 | 18 421 | Apache 2.0 |
| 260 | Llama 3.3 Nemotron Super 49B v1 | NVIDIA | 1 253 | 1 235–1 271 | 968 | Nvidia |
| 261 | DeepSeek V2.5 · v2.5 | DeepSeek | 1 253 | 1 247–1 259 | 11 727 | DeepSeek |
| 262 | Llama 4 Scout | Meta | 1 253 | 1 247–1 259 | 16 748 | Llama |
| 263 | Qwen2.5 72B Instruct | Alibaba Qwen | 1 252 | 1 246–1 257 | 18 845 | Qwen |
| 264 | Llama 3.1 405B Instruct · fp8 | Meta | 1 248 | 1 243–1 253 | 27 214 | Llama 3.1 Community |
| 265 | Llama 3.1 405B Instruct · bf16 | Meta | 1 247 | 1 242–1 252 | 18 057 | Llama 3.1 Community |
| 266 | GPT-4 Turbo Preview · 1106-preview | OpenAI | 1 246 | 1 240–1 251 | 35 407 | Proprietary |
| 267 | GPT-4 Turbo | OpenAI | 1 245 | 1 240–1 250 | 42 998 | Proprietary |
| 268 | Llama 3.1 Nemotron 70B Instruct | NVIDIA | 1 245 | 1 235–1 255 | 3 426 | Llama 3.1 |
| 269 | GPT-4 Turbo Preview · 0125-preview | OpenAI | 1 241 | 1 235–1 246 | 38 472 | Proprietary |
| 270 | Claude 3.5 Haiku | Anthropic | 1 237 | 1 233–1 242 | 33 507 | Proprietary |
| 271 | Reka Core | Reka AI | 1 236 | 1 226–1 246 | 3 378 | Proprietary |
| 272 | Mistral Large 2407 | Mistral AI | 1 236 | 1 231–1 242 | 20 596 | Mistral Research |
| 273 | Llama 3.3 70B Instruct | Meta | 1 236 | 1 231–1 240 | 25 201 | Llama-3.3 |
| 274 | Hunyuan Large Vision | Tencent | 1 236 | 1 224–1 248 | 2 768 | Proprietary |
| 275 | Llama 3.1 Tulu 3 70B | Ai2 | 1 235 | 1 220–1 251 | 1 217 | Llama 3.1 |
| 276 | Mistral Large | Mistral AI | 1 235 | 1 229–1 241 | 11 784 | MRL |
| 277 | Athene 70B | Nexusflow | 1 235 | 1 227–1 242 | 8 353 | CC-BY-NC-4.0 |
| 278 | Nova Pro 1.0 | Amazon | 1 233 | 1 227–1 239 | 10 557 | Proprietary |
| 279 | Magistral Medium | Mistral AI | 1 232 | 1 224–1 240 | 7 030 | Proprietary |
| 280 | Gemini 1.5 Flash · 001 | 1 230 | 1 225–1 236 | 29 023 | Proprietary | |
| 281 | Llama 3.1 70B Instruct | Meta | 1 219 | 1 214–1 224 | 25 546 | Llama 3.1 Community |
| 282 | Gemma 2 27B | 1 217 | 1 212–1 221 | 34 471 | Gemma license | |
| 283 | Command R+ (08-2024) | Cohere | 1 216 | 1 207–1 225 | 4 561 | CC-BY-NC-4.0 |
| 284 | Gemini 1.5 Flash-8B | 1 214 | 1 209–1 220 | 17 091 | Proprietary | |
| 285 | Granite 4.0 H Small | IBM Granite | 1 214 | 1 202–1 225 | 3 098 | Apache 2.0 |
| 286 | Aya Expanse 32B | Cohere | 1 213 | 1 207–1 220 | 12 766 | CC-BY-NC-4.0 |
| 287 | OLMo 3.1 32B Think | Ai2 | 1 210 | 1 200–1 219 | 4 897 | Apache 2.0 |
| 288 | Nova Lite 1.0 | Amazon | 1 208 | 1 201–1 215 | 8 328 | Proprietary |
| 289 | Reka Flash (2024-09) | Reka AI | 1 207 | 1 197–1 217 | 3 504 | Proprietary |
| 290 | Gemma 2 9B IT SimPO | Princeton NLP | 1 206 | 1 197–1 216 | 4 238 | MIT |
| 291 | Nemotron-4 340B Instruct | NVIDIA | 1 206 | 1 199–1 213 | 9 367 | NVIDIA Open Model |
| 292 | Claude 3 Sonnet | Anthropic | 1 205 | 1 200–1 210 | 48 314 | Proprietary |
| 293 | Qwen2.5 Coder 32B Instruct | Alibaba Qwen | 1 204 | 1 194–1 215 | 2 780 | Apache 2.0 |
| 294 | GLM-4 | Z.ai | 1 200 | 1 190–1 209 | 4 619 | Proprietary |
| 295 | Mistral Small 3 | Mistral AI | 1 198 | 1 190–1 206 | 5 941 | Apache 2.0 |
| 296 | Phi 4 | Microsoft | 1 197 | 1 190–1 203 | 10 025 | MIT |
| 297 | Jamba 1.5 Large | AI21 Labs | 1 196 | 1 186–1 206 | 3 747 | Jamba Open |
| 298 | GPT-4 | OpenAI | 1 194 | 1 187–1 201 | 18 639 | Proprietary |
| 299 | Command R+ | Cohere | 1 189 | 1 184–1 195 | 35 178 | CC-BY-NC-4.0 |
| 300 | Gemma 2 9B | 1 188 | 1 183–1 193 | 24 824 | Gemma license |
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.