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: право, госуправление», 300 строк. Оценка и интервал — из источника; интервал не равен диапазону мест.
| Место | Модель | Разработчик | Оценка | Интервал | Голоса | Лицензия |
|---|---|---|---|---|---|---|
| 1 | Claude Opus 5 в режиме max | Anthropic | 1 534 | 1 520–1 549 | 1 755 | Proprietary |
| 2 | Muse Spark 1.2 в режиме xhigh | Meta | 1 532 | 1 498–1 566 | 273 | Proprietary |
| 3 | Claude Opus 5 в режиме high | Anthropic | 1 526 | 1 515–1 537 | 3 594 | Proprietary |
| 4 | Claude Fable 5.1 в режиме max | Anthropic | 1 520 | 1 492–1 548 | 452 | Proprietary |
| 5 | Claude Opus 4.6 в режиме high | Anthropic | 1 512 | 1 504–1 521 | 5 876 | Proprietary |
| 6 | Claude Opus 4.6 | Anthropic | 1 508 | 1 499–1 516 | 6 134 | Proprietary |
| 7 | Claude Opus 4.7 в режиме high | Anthropic | 1 505 | 1 496–1 515 | 4 832 | Proprietary |
| 8 | Kimi K3 в режиме max | Moonshot AI | 1 502 | 1 487–1 517 | 1 682 | Kimi K3 license |
| 9 | Gemini 3.7 Flash в режиме high | 1 500 | 1 471–1 528 | 436 | Proprietary | |
| 10 | Claude Fable 5 | Anthropic | 1 500 | 1 487–1 512 | 2 509 | Proprietary |
| 11 | Muse Spark 1.3 в режиме max | Meta | 1 497 | 1 469–1 525 | 406 | Proprietary |
| 12 | Gemini 3 Pro | 1 492 | 1 481–1 503 | 2 884 | Proprietary | |
| 13 | Gemini 3.5 Flash в режиме medium | 1 490 | 1 479–1 502 | 2 960 | Proprietary | |
| 14 | GPT-5.5 в режиме high | OpenAI | 1 490 | 1 481–1 499 | 5 286 | Proprietary |
| 15 | Muse Spark | Meta | 1 489 | 1 469–1 508 | 973 | Proprietary |
| 16 | Gemini 3.8 Flash в режиме high | 1 488 | 1 461–1 514 | 494 | Proprietary | |
| 17 | Claude Opus 4.7 | Anthropic | 1 487 | 1 478–1 496 | 5 039 | Proprietary |
| 18 | GLM 5.3 в режиме max | Z.ai | 1 486 | 1 467–1 506 | 984 | MIT |
| 19 | GPT-5.4 в режиме high | OpenAI | 1 486 | 1 477–1 495 | 4 885 | Proprietary |
| 20 | Gemini 3.1 Pro Preview | 1 486 | 1 479–1 493 | 8 809 | Proprietary | |
| 21 | GPT-6 Astra в режиме max | OpenAI | 1 484 | 1 445–1 523 | 213 | Proprietary |
| 22 | Qwen3.5 Max | Alibaba Qwen | 1 484 | 1 469–1 498 | 1 669 | Proprietary |
| 23 | Gemini 3.5 Flash в режиме high | 1 483 | 1 472–1 495 | 3 187 | Proprietary | |
| 24 | Qwen3.8 Max (0902) | Alibaba Qwen | 1 483 | 1 466–1 500 | 1 240 | Proprietary |
| 25 | Muse Spark 1.1 | Meta | 1 483 | 1 470–1 495 | 2 393 | Proprietary |
| 26 | Gemini 3.6 Flash в режиме high | 1 482 | 1 469–1 495 | 2 226 | Proprietary | |
| 27 | GPT-5.5 · 5.5 | OpenAI | 1 482 | 1 473–1 491 | 5 473 | Proprietary |
| 28 | Gemini 2.5 Pro | 1 479 | 1 473–1 486 | 8 714 | Proprietary | |
| 29 | Gemini 3 Flash Preview | 1 475 | 1 463–1 488 | 2 300 | Proprietary | |
| 30 | MiMo-V2.5-Pro | Xiaomi | 1 475 | 1 466–1 484 | 4 938 | MIT |
| 31 | Claude Opus 4.8 в режиме high | Anthropic | 1 475 | 1 465–1 484 | 4 450 | Proprietary |
| 32 | Claude Opus 4.8 | Anthropic | 1 474 | 1 464–1 484 | 4 488 | Proprietary |
| 33 | GLM 5.2 в режиме max | Z.ai | 1 473 | 1 462–1 485 | 2 995 | MIT |
| 34 | Qwen3.7 Max | Alibaba Qwen | 1 472 | 1 438–1 507 | 290 | Proprietary |
| 35 | Claude Sonnet 4.6 | Anthropic | 1 471 | 1 462–1 480 | 5 366 | Proprietary |
| 36 | GPT-5.6 Sol в режиме xhigh | OpenAI | 1 471 | 1 458–1 484 | 2 270 | Proprietary |
| 37 | ERNIE 5.1 | Baidu | 1 469 | 1 457–1 480 | 2 931 | Proprietary |
| 38 | GPT-5.6 Terra в режиме xhigh | OpenAI | 1 469 | 1 456–1 481 | 2 384 | Proprietary |
| 39 | GLM 5.3 Flash | Z.ai | 1 468 | 1 448–1 488 | 853 | MIT |
| 40 | GLM 5.1 | Z.ai | 1 467 | 1 457–1 477 | 3 853 | MIT |
| 41 | GPT-5.4 | OpenAI | 1 464 | 1 455–1 473 | 5 045 | Proprietary |
| 42 | DeepSeek V4 Pro 0423 | DeepSeek | 1 463 | 1 454–1 473 | 4 229 | MIT |
| 43 | ERNIE 5.0 · preview-1022 | Baidu | 1 463 | 1 433–1 493 | 319 | Proprietary |
| 44 | Amazon Nova Experimental Chat · 26-02-10 | Amazon | 1 462 | 1 426–1 498 | 250 | Proprietary |
| 45 | DeepSeek V4 Pro 0813 в режиме high | DeepSeek | 1 460 | 1 439–1 482 | 748 | MIT |
| 46 | Claude Opus 4.5 | Anthropic | 1 459 | 1 451–1 468 | 5 188 | Proprietary |
| 47 | Grok 4.5 | xAI | 1 459 | 1 447–1 472 | 2 584 | Proprietary |
| 48 | DeepSeek V3.1 Terminus · terminus-thinking | DeepSeek | 1 459 | 1 424–1 495 | 257 | MIT |
| 49 | Kimi K2.6 | Moonshot AI | 1 459 | 1 448–1 471 | 2 844 | Modified MIT |
| 50 | Nemotron 3 Ultra | NVIDIA | 1 458 | 1 439–1 478 | 907 | OpenMDW-1.1 |
| 51 | Claude Sonnet 5 в режиме high | Anthropic | 1 458 | 1 447–1 469 | 3 064 | Proprietary |
| 52 | DeepSeek V4 Pro 0423 в режиме high | DeepSeek | 1 458 | 1 448–1 468 | 4 054 | MIT |
| 53 | GPT-5.1 в режиме high | OpenAI | 1 458 | 1 446–1 469 | 2 939 | Proprietary |
| 54 | Gemma 4 26B A4B | 1 456 | 1 428–1 484 | 390 | Apache 2.0 | |
| 55 | Qwen3.6 Max Preview | Alibaba Qwen | 1 455 | 1 424–1 485 | 384 | Proprietary |
| 56 | GLM 5 | Z.ai | 1 454 | 1 441–1 467 | 2 125 | MIT |
| 57 | Claude Opus 4.5 в режиме high | Anthropic | 1 454 | 1 442–1 465 | 2 634 | Proprietary |
| 58 | Grok 4.20 · beta-0309-reasoning | xAI | 1 453 | 1 444–1 463 | 4 941 | Proprietary |
| 59 | Qwen3.7 Plus | Alibaba Qwen | 1 451 | 1 441–1 462 | 3 344 | Proprietary |
| 60 | Claude Sonnet 4.5 | Anthropic | 1 450 | 1 442–1 458 | 5 727 | Proprietary |
| 61 | Inkling | Thinking Machines Lab | 1 449 | 1 436–1 462 | 2 122 | Apache 2.0 |
| 62 | Grok 4.20 Multi-Agent | xAI | 1 449 | 1 440–1 458 | 4 861 | Proprietary |
| 63 | Qwen3.8 27B | Alibaba Qwen | 1 448 | 1 427–1 469 | 827 | Apache 2.0 |
| 64 | Seed 2.0 Pro | ByteDance Seed | 1 448 | 1 439–1 456 | 5 756 | Proprietary |
| 65 | GLM 4.6 | Z.ai | 1 448 | 1 436–1 460 | 2 367 | MIT |
| 66 | GPT-5.6 Luna в режиме xhigh | OpenAI | 1 447 | 1 435–1 460 | 2 394 | Proprietary |
| 67 | ERNIE 5.0 · preview-1203 | Baidu | 1 447 | 1 426–1 468 | 757 | Proprietary |
| 68 | MiMo-V2 Pro | Xiaomi | 1 446 | 1 432–1 461 | 1 755 | Proprietary |
| 69 | Gemini 3 Flash Preview в режиме minimal | 1 446 | 1 438–1 454 | 6 829 | Proprietary | |
| 70 | Grok 3 | xAI | 1 445 | 1 431–1 459 | 1 869 | Proprietary |
| 71 | ChatGPT-4o (latest) | OpenAI | 1 444 | 1 436–1 453 | 5 171 | Proprietary |
| 72 | GLM 4.7 | Z.ai | 1 444 | 1 425–1 463 | 954 | MIT |
| 73 | Qwen3 VL 235B A22B Instruct | Alibaba Qwen | 1 444 | 1 421–1 466 | 732 | Apache 2.0 |
| 74 | Grok 4.1 · 4.1 | xAI | 1 443 | 1 434–1 452 | 4 844 | Proprietary |
| 75 | Kimi K2.5 · thinking | Moonshot AI | 1 441 | 1 433–1 450 | 5 320 | Modified MIT |
| 76 | Gemma 4 31B | 1 441 | 1 414–1 468 | 437 | Apache 2.0 | |
| 77 | Amazon Nova Experimental Chat · 12-10 | Amazon | 1 440 | 1 405–1 476 | 253 | Proprietary |
| 78 | DeepSeek V4 Flash 0423 в режиме high | DeepSeek | 1 440 | 1 430–1 450 | 3 871 | MIT |
| 79 | DeepSeek V4 Flash 0423 | DeepSeek | 1 440 | 1 430–1 450 | 3 886 | MIT |
| 80 | Claude Sonnet 4.5 в режиме high | Anthropic | 1 440 | 1 432–1 448 | 5 616 | Proprietary |
| 81 | Grok 4.1 · 4.1-thinking | xAI | 1 439 | 1 430–1 448 | 4 721 | Proprietary |
| 82 | Mistral Large 3 2512 | Mistral AI | 1 439 | 1 431–1 448 | 5 184 | Apache 2.0 |
| 83 | Mistral Medium 3.1 | Mistral AI | 1 439 | 1 432–1 447 | 6 482 | Proprietary |
| 84 | GPT-5.1 | OpenAI | 1 439 | 1 429–1 450 | 3 050 | Proprietary |
| 85 | Qwen3.6 Plus | Alibaba Qwen | 1 438 | 1 428–1 449 | 3 433 | Proprietary |
| 86 | Qwen3.5 397B A17B | Alibaba Qwen | 1 438 | 1 430–1 447 | 5 845 | Apache 2.0 |
| 87 | GPT-5.2 Chat | OpenAI | 1 437 | 1 425–1 449 | 2 580 | Proprietary |
| 88 | GLM 4.5 | Z.ai | 1 437 | 1 421–1 452 | 1 461 | MIT |
| 89 | Grok 4.20 · beta1 | xAI | 1 436 | 1 422–1 450 | 2 018 | Proprietary |
| 90 | ERNIE 5.0 · 0110 | Baidu | 1 436 | 1 424–1 448 | 2 545 | Proprietary |
| 91 | Grok 4.6 в режиме high | xAI | 1 436 | 1 420–1 452 | 1 434 | Proprietary |
| 92 | Mistral Medium 3.5 | Mistral AI | 1 436 | 1 415–1 456 | 837 | Modified MIT |
| 93 | Qwen3 Max · preview | Alibaba Qwen | 1 435 | 1 421–1 449 | 1 688 | Proprietary |
| 94 | GPT-4.5 Preview | OpenAI | 1 434 | 1 411–1 456 | 701 | Proprietary |
| 95 | MiniMax M3 | MiniMax | 1 433 | 1 423–1 443 | 3 983 | MiniMax Community License |
| 96 | GPT-5 | OpenAI | 1 432 | 1 418–1 445 | 1 893 | Proprietary |
| 97 | Qwen3 Next 80B A3B Instruct | Alibaba Qwen | 1 432 | 1 416–1 447 | 1 386 | Apache 2.0 |
| 98 | Gemini 2.5 Flash · flash | 1 431 | 1 425–1 438 | 8 676 | Proprietary | |
| 99 | GLM 5V Turbo | Z.ai | 1 431 | 1 409–1 453 | 731 | Proprietary |
| 100 | Qwen3 235B A22B Instruct 2507 | Alibaba Qwen | 1 429 | 1 422–1 437 | 6 426 | Apache 2.0 |
| 101 | Grok 4 Fast · reasoning | xAI | 1 429 | 1 411–1 446 | 1 129 | Proprietary |
| 102 | Grok 4 | xAI | 1 428 | 1 417–1 440 | 2 565 | Proprietary |
| 103 | o3 | OpenAI | 1 428 | 1 418–1 439 | 3 731 | Proprietary |
| 104 | Gemini 2.5 Flash · flash-preview-09-2025 | 1 428 | 1 416–1 441 | 2 110 | Proprietary | |
| 105 | Amazon Nova Experimental Chat · 11-10 | Amazon | 1 427 | 1 414–1 441 | 1 818 | Proprietary |
| 106 | Gemini 3.5 Flash Lite | 1 427 | 1 414–1 441 | 2 216 | Proprietary | |
| 107 | Hy3 preview | Tencent | 1 427 | 1 401–1 453 | 532 | tencent-hunyuan-community |
| 108 | GPT-5.2 в режиме high | OpenAI | 1 427 | 1 417–1 437 | 3 567 | Proprietary |
| 109 | MiMo-V2 Omni | Xiaomi | 1 426 | 1 411–1 441 | 1 606 | Proprietary |
| 110 | DeepSeek V3.2 · v3.2 | DeepSeek | 1 426 | 1 416–1 436 | 3 360 | MIT |
| 111 | Kimi K2.5 · instant | Moonshot AI | 1 426 | 1 401–1 451 | 513 | Modified MIT |
| 112 | Hy3 | Tencent | 1 425 | 1 401–1 450 | 620 | Apache 2.0 |
| 113 | DeepSeek V3.2 Exp · exp-thinking | DeepSeek | 1 424 | 1 399–1 449 | 493 | MIT |
| 114 | Qwen3.5-122B-A10B | Alibaba Qwen | 1 423 | 1 410–1 435 | 2 227 | Apache 2.0 |
| 115 | GPT-5 в режиме high | OpenAI | 1 422 | 1 408–1 436 | 1 823 | Proprietary |
| 116 | DeepSeek V3.2 · v3.2-thinking | DeepSeek | 1 422 | 1 411–1 432 | 2 961 | MIT |
| 117 | MiMo-V2.5 | Xiaomi | 1 421 | 1 411–1 432 | 3 570 | MIT |
| 118 | GPT-5.2 | OpenAI | 1 421 | 1 413–1 429 | 6 304 | Proprietary |
| 119 | GPT-5.5 · 5.5-instant | OpenAI | 1 420 | 1 407–1 434 | 2 183 | Proprietary |
| 120 | R1 0528 | DeepSeek | 1 419 | 1 403–1 436 | 1 371 | MIT |
| 121 | Claude Opus 4.1 · 20250805-thinking-16k | Anthropic | 1 419 | 1 408–1 430 | 3 148 | Proprietary |
| 122 | DeepSeek V3.1 · v3.1-thinking | DeepSeek | 1 418 | 1 397–1 439 | 772 | MIT |
| 123 | Gemini 3.1 Flash Lite Preview | 1 418 | 1 409–1 427 | 4 742 | Proprietary | |
| 124 | LongCat-Flash-Chat · chat | Meituan | 1 418 | 1 397–1 439 | 774 | MIT |
| 125 | Kimi K2 Thinking | Moonshot AI | 1 417 | 1 408–1 426 | 4 404 | Modified MIT |
| 126 | Inkling Small | Thinking Machines Lab | 1 417 | 1 402–1 432 | 1 641 | Apache 2.0 |
| 127 | Claude Opus 4.1 · 20250805 | Anthropic | 1 417 | 1 409–1 426 | 5 206 | Proprietary |
| 128 | Hunyuan T1 | Tencent | 1 417 | 1 383–1 450 | 273 | Proprietary |
| 129 | LongCat-Flash-Chat · chat-2602-exp | Meituan | 1 416 | 1 403–1 429 | 2 101 | Proprietary |
| 130 | DeepSeek V3.1 · v3.1 | DeepSeek | 1 416 | 1 397–1 434 | 1 000 | MIT |
| 131 | Qwen3 235B A22B Thinking 2507 | Alibaba Qwen | 1 415 | 1 391–1 440 | 531 | Apache 2.0 |
| 132 | GPT-5.4 Mini в режиме high | OpenAI | 1 414 | 1 405–1 424 | 4 730 | Proprietary |
| 133 | MiMo-V2 Flash · flash (non-thinking) | Xiaomi | 1 412 | 1 402–1 423 | 3 337 | MIT |
| 134 | MiniMax M2.1 | MiniMax | 1 411 | 1 394–1 428 | 1 157 | MIT |
| 135 | Grok 4 Fast · chat | xAI | 1 409 | 1 381–1 438 | 401 | Proprietary |
| 136 | DeepSeek V3.2 Exp · exp | DeepSeek | 1 409 | 1 389–1 430 | 780 | MIT |
| 137 | Step 3.5 Flash | StepFun | 1 408 | 1 399–1 417 | 4 412 | Apache 2.0 |
| 138 | Qwen3 VL 235B A22B Thinking | Alibaba Qwen | 1 407 | 1 382–1 433 | 490 | Apache 2.0 |
| 139 | Qwen3.5-27B | Alibaba Qwen | 1 407 | 1 394–1 420 | 2 095 | Apache 2.0 |
| 140 | DeepSeek V3.1 Terminus · terminus | DeepSeek | 1 405 | 1 370–1 441 | 264 | MIT |
| 141 | MiniMax M2.7 | MiniMax | 1 403 | 1 394–1 412 | 5 294 | Modified MIT |
| 142 | Kimi K2 0905 | Moonshot AI | 1 403 | 1 382–1 424 | 769 | Modified MIT |
| 143 | Qwen3 Max · 2025-09-23 | Alibaba Qwen | 1 401 | 1 376–1 426 | 541 | Proprietary |
| 144 | MiMo-V2 Flash · flash (thinking) | Xiaomi | 1 401 | 1 381–1 421 | 841 | MIT |
| 145 | Grok 4.1 Fast | xAI | 1 401 | 1 391–1 410 | 4 043 | Proprietary |
| 146 | Claude Haiku 4.5 | Anthropic | 1 400 | 1 394–1 407 | 9 611 | Proprietary |
| 147 | GPT-4.1 | OpenAI | 1 399 | 1 388–1 409 | 3 214 | Proprietary |
| 148 | GLM 4.6V | Z.ai | 1 399 | 1 357–1 440 | 200 | MIT |
| 149 | Qwen3.5-Flash | Alibaba Qwen | 1 398 | 1 389–1 408 | 4 531 | Proprietary |
| 150 | Qwen3.5-35B-A3B | Alibaba Qwen | 1 398 | 1 385–1 410 | 2 282 | Apache 2.0 |
| 151 | Qwen3 235B A22B · a22b-no-thinking | Alibaba Qwen | 1 395 | 1 383–1 407 | 2 524 | Apache 2.0 |
| 152 | Gemini 2.5 Flash Lite · 09-2025-no-thinking | 1 393 | 1 382–1 403 | 3 088 | Proprietary | |
| 153 | Nemotron 3 Super | NVIDIA | 1 393 | 1 367–1 418 | 520 | NVIDIA Open Model |
| 154 | GPT-5.3 Chat | OpenAI | 1 392 | 1 380–1 404 | 2 554 | Proprietary |
| 155 | Grok 3 Mini в режиме high | xAI | 1 390 | 1 372–1 408 | 1 087 | Proprietary |
| 156 | Amazon Nova Experimental Chat · 10-20 | Amazon | 1 388 | 1 368–1 409 | 817 | Proprietary |
| 157 | Qwen3 30B A3B Instruct 2507 | Alibaba Qwen | 1 387 | 1 371–1 402 | 1 447 | Apache 2.0 |
| 158 | Muse Glimmer 30B | Meta | 1 384 | 1 349–1 418 | 286 | Apache-2.0 |
| 159 | Mistral Medium 3 | Mistral AI | 1 384 | 1 371–1 396 | 2 116 | Proprietary |
| 160 | GLM 4.5 Air | Z.ai | 1 383 | 1 370–1 397 | 1 891 | MIT |
| 161 | Grok 4.3 | xAI | 1 382 | 1 373–1 391 | 5 431 | Proprietary |
| 162 | Gemma 3 12B | 1 381 | 1 343–1 419 | 217 | Gemma | |
| 163 | DeepSeek V3 0324 | DeepSeek | 1 380 | 1 369–1 392 | 2 896 | MIT |
| 164 | Amazon Nova Experimental Chat · 26-01-10 | Amazon | 1 378 | 1 343–1 413 | 262 | Proprietary |
| 165 | Qwen2.5 Max | Alibaba Qwen | 1 376 | 1 361–1 390 | 1 856 | Proprietary |
| 166 | GPT-5 Mini в режиме high | OpenAI | 1 375 | 1 361–1 390 | 1 683 | Proprietary |
| 167 | Gemma 3 27B | 1 375 | 1 363–1 386 | 2 653 | Gemma | |
| 168 | Claude Opus 4 · 20250514 | Anthropic | 1 374 | 1 363–1 385 | 2 847 | Proprietary |
| 169 | Gemini 2.5 Flash Lite · 06-17-thinking | 1 372 | 1 359–1 385 | 2 054 | Proprietary | |
| 170 | Grok 3 Mini | xAI | 1 372 | 1 356–1 388 | 1 408 | Proprietary |
| 171 | Hunyuan TurboS | Tencent | 1 372 | 1 350–1 393 | 736 | Proprietary |
| 172 | R1 | DeepSeek | 1 371 | 1 354–1 389 | 1 097 | MIT |
| 173 | Gemini 2.0 Flash | 1 371 | 1 359–1 383 | 2 560 | Proprietary | |
| 174 | o1 · 2024-12-17 | OpenAI | 1 368 | 1 353–1 384 | 1 454 | Proprietary |
| 175 | GPT-5.4 Nano в режиме high | OpenAI | 1 368 | 1 358–1 377 | 4 615 | Proprietary |
| 176 | MiniMax M2.5 | MiniMax | 1 367 | 1 355–1 378 | 3 142 | Modified MIT |
| 177 | Claude Opus 4 · 20250514-thinking-16k | Anthropic | 1 366 | 1 353–1 378 | 2 421 | Proprietary |
| 178 | Amazon Nova Experimental Chat · 10-09 | Amazon | 1 365 | 1 327–1 404 | 205 | Proprietary |
| 179 | Qwen3 Next 80B A3B Thinking | Alibaba Qwen | 1 364 | 1 344–1 384 | 838 | Apache 2.0 |
| 180 | GLM-4-Plus · plus-0111 | Z.ai | 1 362 | 1 332–1 392 | 367 | Proprietary |
| 181 | Ling-flash-2.0 | inclusionAI | 1 362 | 1 334–1 389 | 413 | MIT |
| 182 | GLM 4.7 Flash | Z.ai | 1 360 | 1 340–1 380 | 815 | MIT |
| 183 | Qwen3 Coder 480B A35B | Alibaba Qwen | 1 359 | 1 343–1 374 | 1 561 | Apache 2.0 |
| 184 | Qwen3 235B A22B · a22b | Alibaba Qwen | 1 356 | 1 342–1 370 | 1 714 | Apache 2.0 |
| 185 | gpt-oss-120b | OpenAI | 1 355 | 1 341–1 369 | 1 859 | Apache 2.0 |
| 186 | Nova 2 Lite | Amazon | 1 355 | 1 335–1 375 | 878 | Proprietary |
| 187 | Kimi K2 0711 | Moonshot AI | 1 355 | 1 340–1 369 | 1 747 | Modified MIT |
| 188 | Step-1o Turbo | StepFun | 1 354 | 1 331–1 377 | 653 | Proprietary |
| 189 | o4 Mini | OpenAI | 1 350 | 1 339–1 362 | 2 738 | Proprietary |
| 190 | Granite 4.2 30B | IBM Granite | 1 349 | 1 312–1 387 | 246 | Apache 2.0 |
| 191 | Qwen-Plus | Alibaba Qwen | 1 349 | 1 320–1 377 | 392 | Proprietary |
| 192 | Command A | Cohere | 1 348 | 1 338–1 359 | 3 415 | CC-BY-NC-4.0 |
| 193 | MiniMax M2 | MiniMax | 1 348 | 1 320–1 376 | 424 | Apache 2.0 |
| 194 | Trinity Large Thinking | Arcee AI | 1 347 | 1 334–1 360 | 2 301 | Apache 2.0 |
| 195 | o1 · preview | OpenAI | 1 346 | 1 332–1 361 | 1 945 | Proprietary |
| 196 | Solar Pro 4 | Upstage | 1 346 | 1 304–1 388 | 207 | Proprietary |
| 197 | Nemotron 3 Nano 30B A3B | NVIDIA | 1 345 | 1 328–1 362 | 1 189 | NVIDIA Open Model |
| 198 | Llama 3.3 Nemotron Super 49B v1.5 | NVIDIA | 1 345 | 1 307–1 382 | 232 | Nvidia Open |
| 199 | Gemini 2.0 Flash-Lite | 1 345 | 1 329–1 360 | 1 435 | Proprietary | |
| 200 | Trinity Large | Arcee AI | 1 344 | 1 330–1 358 | 2 144 | Apache 2.0 |
| 201 | GPT-4.1 Mini | OpenAI | 1 344 | 1 332–1 356 | 2 439 | Proprietary |
| 202 | Claude Sonnet 4 · 20250514 | Anthropic | 1 342 | 1 330–1 354 | 2 606 | Proprietary |
| 203 | DeepSeek V3 | DeepSeek | 1 341 | 1 325–1 358 | 1 312 | DeepSeek |
| 204 | Mistral Small 3.2 24B | Mistral AI | 1 341 | 1 324–1 358 | 1 147 | Apache 2.0 |
| 205 | GLM 4.5V | Z.ai | 1 339 | 1 305–1 373 | 270 | MIT |
| 206 | MiniMax M1 | MiniMax | 1 339 | 1 326–1 351 | 2 331 | Apache 2.0 |
| 207 | Step 3 | StepFun | 1 338 | 1 310–1 367 | 407 | Apache 2.0 |
| 208 | Claude Sonnet 4 · 20250514-thinking-32k | Anthropic | 1 337 | 1 324–1 350 | 2 201 | Proprietary |
| 209 | Step-2 16k | StepFun | 1 337 | 1 303–1 370 | 278 | Proprietary |
| 210 | QwQ 32B | Alibaba Qwen | 1 336 | 1 322–1 351 | 1 621 | Apache 2.0 |
| 211 | OLMo 3.1 32B Instruct | Ai2 | 1 331 | 1 311–1 351 | 847 | Apache 2.0 |
| 212 | o3 Mini High | OpenAI | 1 329 | 1 311–1 347 | 1 023 | Proprietary |
| 213 | Gemini 1.5 Pro · 1.5-pro-002 | 1 328 | 1 317–1 340 | 3 342 | Proprietary | |
| 214 | Yi-Lightning | 01.AI | 1 328 | 1 313–1 343 | 1 697 | Proprietary |
| 215 | Gemma 3 4B | 1 327 | 1 292–1 363 | 245 | Gemma | |
| 216 | Granite 4.1 8B | IBM Granite | 1 327 | 1 286–1 368 | 236 | Apache 2.0 |
| 217 | Grok 2 | xAI | 1 326 | 1 315–1 337 | 3 813 | Proprietary |
| 218 | Qwen3 32B | Alibaba Qwen | 1 325 | 1 293–1 357 | 285 | Apache 2.0 |
| 219 | Granite 4.2 8B | IBM Granite | 1 324 | 1 286–1 363 | 256 | Apache 2.0 |
| 220 | INTELLECT-3 | Prime Intellect | 1 323 | 1 294–1 352 | 399 | MIT |
| 221 | Granite 4.2 3B | IBM Granite | 1 315 | 1 275–1 356 | 248 | Apache 2.0 |
| 222 | o1-mini | OpenAI | 1 315 | 1 304–1 327 | 3 138 | Proprietary |
| 223 | Qwen3 30B A3B | Alibaba Qwen | 1 314 | 1 300–1 328 | 1 752 | Apache 2.0 |
| 224 | GPT-4o (2024-05-13) | OpenAI | 1 314 | 1 304–1 324 | 6 743 | Proprietary |
| 225 | GLM-4-Plus · plus | Z.ai | 1 309 | 1 294–1 324 | 1 723 | Proprietary |
| 226 | Gemma 3n E4B | 1 309 | 1 293–1 325 | 1 386 | Gemma | |
| 227 | Nemotron 3.5 Lightning | NVIDIA | 1 308 | 1 284–1 333 | 666 | OpenMDW-1.1 |
| 228 | Ring-flash-2.0 | inclusionAI | 1 308 | 1 279–1 337 | 422 | MIT |
| 229 | Qwen2.5 Plus | Alibaba Qwen | 1 307 | 1 283–1 331 | 588 | Proprietary |
| 230 | GPT-4o-mini (2024-07-18) | OpenAI | 1 305 | 1 295–1 316 | 3 866 | Proprietary |
| 231 | o3 Mini | OpenAI | 1 305 | 1 295–1 315 | 3 605 | Proprietary |
| 232 | Gemini 1.5 Pro · advanced-0514 | 1 304 | 1 290–1 318 | 2 962 | Proprietary | |
| 233 | Llama 3.1 405B Instruct · fp8 | Meta | 1 303 | 1 292–1 315 | 3 383 | Llama 3.1 Community |
| 234 | Llama 3.1 Nemotron 70B Instruct | NVIDIA | 1 303 | 1 277–1 330 | 489 | Llama 3.1 |
| 235 | Gemini 1.5 Flash · 002 | 1 302 | 1 289–1 315 | 2 160 | Proprietary | |
| 236 | Claude 3.7 Sonnet · 20250219-thinking-32k | Anthropic | 1 302 | 1 290–1 314 | 2 469 | Proprietary |
| 237 | Llama 3.1 405B Instruct · bf16 | Meta | 1 302 | 1 289–1 315 | 2 330 | Llama 3.1 Community |
| 238 | Claude 3.7 Sonnet · 20250219 | Anthropic | 1 300 | 1 288–1 313 | 2 476 | Proprietary |
| 239 | GPT-4.1 Nano | OpenAI | 1 300 | 1 272–1 328 | 382 | Proprietary |
| 240 | Athene V2 Chat | Nexusflow | 1 299 | 1 283–1 315 | 1 443 | NexusFlow |
| 241 | GPT-5 Nano в режиме high | OpenAI | 1 298 | 1 271–1 325 | 469 | Proprietary |
| 242 | Athene 70B | Nexusflow | 1 293 | 1 274–1 312 | 1 008 | CC-BY-NC-4.0 |
| 243 | Grok 2 Mini | xAI | 1 292 | 1 281–1 304 | 3 012 | Proprietary |
| 244 | Hunyuan Standard | Tencent | 1 290 | 1 256–1 324 | 288 | Proprietary |
| 245 | OLMo 3 32B Think | Ai2 | 1 289 | 1 260–1 319 | 410 | Apache 2.0 |
| 246 | Qwen2.5 72B Instruct | Alibaba Qwen | 1 288 | 1 275–1 301 | 2 386 | Qwen |
| 247 | Llama 4 Maverick | Meta | 1 285 | 1 273–1 297 | 2 591 | Llama 4 |
| 248 | Llama 4 Scout | Meta | 1 285 | 1 271–1 299 | 1 978 | Llama |
| 249 | Llama 3.3 70B Instruct | Meta | 1 284 | 1 273–1 295 | 3 150 | Llama-3.3 |
| 250 | Llama 3.1 70B Instruct | Meta | 1 284 | 1 272–1 295 | 3 177 | Llama 3.1 Community |
| 251 | Qwen Max | Alibaba Qwen | 1 283 | 1 265–1 301 | 1 085 | Qwen |
| 252 | OLMo 3.1 32B Think | Ai2 | 1 282 | 1 258–1 306 | 618 | Apache 2.0 |
| 253 | DeepSeek V2.5 · v2.5 | DeepSeek | 1 282 | 1 267–1 298 | 1 518 | DeepSeek |
| 254 | Nova Pro 1.0 | Amazon | 1 281 | 1 265–1 298 | 1 296 | Proprietary |
| 255 | Hunyuan Large | Tencent | 1 281 | 1 245–1 317 | 252 | Proprietary |
| 256 | gpt-oss-20b | OpenAI | 1 279 | 1 256–1 303 | 681 | Apache 2.0 |
| 257 | DeepSeek V2.5 · v2.5-1210 | DeepSeek | 1 279 | 1 249–1 308 | 375 | DeepSeek |
| 258 | Claude 3.5 Sonnet · 20241022 | Anthropic | 1 277 | 1 268–1 286 | 5 492 | Proprietary |
| 259 | Mistral Large 2407 | Mistral AI | 1 276 | 1 264–1 288 | 2 735 | Mistral Research |
| 260 | GPT-4o (2024-08-06) | OpenAI | 1 275 | 1 263–1 288 | 2 674 | Proprietary |
| 261 | Mistral Large | Mistral AI | 1 275 | 1 260–1 290 | 1 593 | MRL |
| 262 | Reka Core | Reka AI | 1 274 | 1 245–1 302 | 371 | Proprietary |
| 263 | Mistral Small 3.1 24B | Mistral AI | 1 270 | 1 257–1 284 | 2 109 | Apache 2.0 |
| 264 | Gemini 1.5 Pro · 1.5-pro-001 | 1 270 | 1 258–1 282 | 4 795 | Proprietary | |
| 265 | Gemma 2 9B IT SimPO | Princeton NLP | 1 269 | 1 244–1 293 | 543 | MIT |
| 266 | Command R+ (08-2024) | Cohere | 1 268 | 1 245–1 292 | 592 | CC-BY-NC-4.0 |
| 267 | Claude 3.5 Sonnet · 20240620 | Anthropic | 1 266 | 1 255–1 277 | 4 818 | Proprietary |
| 268 | GPT-4 Turbo | OpenAI | 1 265 | 1 254–1 276 | 5 381 | Proprietary |
| 269 | Claude 3 Opus | Anthropic | 1 265 | 1 255–1 274 | 11 095 | Proprietary |
| 270 | Jamba 1.5 Large | AI21 Labs | 1 264 | 1 237–1 291 | 448 | Jamba Open |
| 271 | GPT-4 Turbo Preview · 0125-preview | OpenAI | 1 261 | 1 249–1 273 | 5 037 | Proprietary |
| 272 | Claude 3.5 Haiku | Anthropic | 1 258 | 1 249–1 268 | 4 230 | Proprietary |
| 273 | Qwen2.5 Coder 32B Instruct | Alibaba Qwen | 1 257 | 1 223–1 291 | 273 | Apache 2.0 |
| 274 | Llama 3.1 Nemotron 51B Instruct | NVIDIA | 1 252 | 1 216–1 288 | 230 | Llama 3.1 |
| 275 | GPT-4 Turbo Preview · 1106-preview | OpenAI | 1 249 | 1 238–1 261 | 5 572 | Proprietary |
| 276 | Aya Expanse 32B | Cohere | 1 248 | 1 233–1 262 | 1 742 | CC-BY-NC-4.0 |
| 277 | Gemini 1.5 Flash-8B | 1 248 | 1 234–1 261 | 2 128 | Proprietary | |
| 278 | Gemini 1.5 Flash · 001 | 1 248 | 1 235–1 260 | 3 784 | Proprietary | |
| 279 | Hunyuan Large Vision | Tencent | 1 247 | 1 216–1 278 | 387 | Proprietary |
| 280 | Nemotron-4 340B Instruct | NVIDIA | 1 246 | 1 228–1 264 | 1 177 | NVIDIA Open Model |
| 281 | Command R+ | Cohere | 1 245 | 1 233–1 257 | 4 403 | CC-BY-NC-4.0 |
| 282 | GLM-4 | Z.ai | 1 244 | 1 219–1 269 | 556 | Proprietary |
| 283 | Gemma 2 27B | 1 241 | 1 231–1 252 | 4 458 | Gemma license | |
| 284 | Phi 4 | Microsoft | 1 241 | 1 225–1 258 | 1 341 | MIT |
| 285 | Nova Lite 1.0 | Amazon | 1 238 | 1 220–1 257 | 1 079 | Proprietary |
| 286 | Magistral Medium | Mistral AI | 1 232 | 1 209–1 254 | 843 | Proprietary |
| 287 | Command R (08-2024) | Cohere | 1 230 | 1 206–1 254 | 569 | CC-BY-NC-4.0 |
| 288 | OLMo 2 32B Instruct | Ai2 | 1 228 | 1 188–1 269 | 202 | Apache-2.0 |
| 289 | Llama 3 70B Instruct | Meta | 1 228 | 1 217–1 239 | 8 328 | Llama 3 Community |
| 290 | Aya Expanse 8B | Cohere | 1 225 | 1 200–1 250 | 551 | CC-BY-NC-4.0 |
| 291 | Jamba 1.5 Mini | AI21 Labs | 1 224 | 1 199–1 249 | 531 | Jamba Open |
| 292 | Granite 4.0 H Small | IBM Granite | 1 223 | 1 187–1 259 | 303 | Apache 2.0 |
| 293 | Nova Micro 1.0 | Amazon | 1 221 | 1 203–1 240 | 1 082 | Proprietary |
| 294 | Mistral Small 3 | Mistral AI | 1 219 | 1 198–1 240 | 869 | Apache 2.0 |
| 295 | Reka Flash (2024-09) | Reka AI | 1 217 | 1 188–1 247 | 414 | Proprietary |
| 296 | Gemma 2 9B | 1 215 | 1 204–1 227 | 3 243 | Gemma license | |
| 297 | Claude 3 Sonnet | Anthropic | 1 214 | 1 202–1 226 | 6 062 | Proprietary |
| 298 | Ministral 8B (2410) | Mistral AI | 1 211 | 1 178–1 244 | 324 | MRL |
| 299 | Command R | Cohere | 1 207 | 1 193–1 220 | 3 073 | CC-BY-NC-4.0 |
| 300 | Llama 3.1 8B Instruct | Meta | 1 206 | 1 194–1 219 | 2 882 | Llama 3.1 Community |
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.