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 «французский», 281 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 | 1 539 | 1 493–1 585 | 187 | Proprietary |
| 2 | Claude Opus 5 в режиме high | Anthropic | 1 526 | 1 508–1 543 | 1 354 | Proprietary |
| 3 | Claude Opus 5 в режиме max | Anthropic | 1 517 | 1 492–1 542 | 626 | Proprietary |
| 4 | Claude Opus 4.6 в режиме high | Anthropic | 1 510 | 1 496–1 524 | 2 402 | Proprietary |
| 5 | Qwen3.8 Max (0902) | Alibaba Qwen | 1 509 | 1 484–1 534 | 586 | Proprietary |
| 6 | Claude Fable 5 | Anthropic | 1 505 | 1 485–1 524 | 1 083 | Proprietary |
| 7 | Claude Opus 4.7 в режиме high | Anthropic | 1 504 | 1 489–1 519 | 2 147 | Proprietary |
| 8 | Muse Spark | Meta | 1 502 | 1 474–1 531 | 470 | Proprietary |
| 9 | Gemini 3.6 Flash в режиме high | 1 502 | 1 481–1 523 | 904 | Proprietary | |
| 10 | GLM 5.3 в режиме max | Z.ai | 1 501 | 1 469–1 534 | 339 | MIT |
| 11 | Gemini 3.7 Flash в режиме high | 1 500 | 1 457–1 543 | 209 | Proprietary | |
| 12 | Muse Spark 1.1 | Meta | 1 498 | 1 477–1 518 | 956 | Proprietary |
| 13 | Claude Opus 4.6 | Anthropic | 1 498 | 1 484–1 511 | 2 530 | Proprietary |
| 14 | Claude Opus 4.7 | Anthropic | 1 496 | 1 482–1 511 | 2 258 | Proprietary |
| 15 | GPT-5.4 в режиме high | OpenAI | 1 495 | 1 480–1 509 | 2 286 | Proprietary |
| 16 | Gemini 3 Pro | 1 494 | 1 474–1 514 | 945 | Proprietary | |
| 17 | GLM 5.3 Flash | Z.ai | 1 492 | 1 461–1 523 | 347 | MIT |
| 18 | Gemini 3.1 Pro Preview | 1 488 | 1 476–1 500 | 3 610 | Proprietary | |
| 19 | MiMo-V2.5-Pro | Xiaomi | 1 487 | 1 472–1 501 | 2 282 | MIT |
| 20 | ERNIE 5.1 | Baidu | 1 486 | 1 468–1 503 | 1 461 | Proprietary |
| 21 | Qwen3.5 Max | Alibaba Qwen | 1 485 | 1 462–1 508 | 724 | Proprietary |
| 22 | Kimi K3 в режиме max | Moonshot AI | 1 485 | 1 463–1 508 | 765 | Kimi K3 license |
| 23 | GPT-5.5 · 5.5 | OpenAI | 1 485 | 1 471–1 499 | 2 413 | Proprietary |
| 24 | GPT-5.6 Terra в режиме xhigh | OpenAI | 1 483 | 1 461–1 504 | 876 | Proprietary |
| 25 | GPT-5.5 в режиме high | OpenAI | 1 482 | 1 468–1 496 | 2 407 | Proprietary |
| 26 | Gemini 3 Flash Preview | 1 481 | 1 459–1 503 | 806 | Proprietary | |
| 27 | Gemini 3.5 Flash в режиме high | 1 480 | 1 462–1 499 | 1 258 | Proprietary | |
| 28 | Claude Opus 4.8 в режиме high | Anthropic | 1 480 | 1 464–1 495 | 1 862 | Proprietary |
| 29 | GLM 5.2 в режиме max | Z.ai | 1 479 | 1 462–1 497 | 1 310 | MIT |
| 30 | Nemotron 3 Ultra | NVIDIA | 1 478 | 1 449–1 507 | 449 | OpenMDW-1.1 |
| 31 | GLM 5.1 | Z.ai | 1 477 | 1 461–1 493 | 1 731 | MIT |
| 32 | Claude Opus 4.8 | Anthropic | 1 477 | 1 461–1 492 | 1 950 | Proprietary |
| 33 | Qwen3.7 Plus | Alibaba Qwen | 1 476 | 1 459–1 493 | 1 516 | Proprietary |
| 34 | Gemini 3.5 Flash в режиме medium | 1 476 | 1 457–1 494 | 1 196 | Proprietary | |
| 35 | Grok 4.20 · beta-0309-reasoning | xAI | 1 476 | 1 461–1 490 | 2 119 | Proprietary |
| 36 | Gemini 2.5 Pro | 1 473 | 1 459–1 486 | 2 332 | Proprietary | |
| 37 | GPT-5.6 Sol в режиме xhigh | OpenAI | 1 472 | 1 451–1 493 | 894 | Proprietary |
| 38 | ERNIE 5.0 · preview-1203 | Baidu | 1 472 | 1 434–1 510 | 250 | Proprietary |
| 39 | DeepSeek V4 Pro 0423 | DeepSeek | 1 471 | 1 456–1 486 | 2 118 | MIT |
| 40 | GPT-5.4 | OpenAI | 1 471 | 1 457–1 485 | 2 326 | Proprietary |
| 41 | Claude Opus 4.5 | Anthropic | 1 471 | 1 456–1 486 | 1 966 | Proprietary |
| 42 | Seed 2.0 Pro | ByteDance Seed | 1 470 | 1 456–1 483 | 2 477 | Proprietary |
| 43 | MiMo-V2 Pro | Xiaomi | 1 469 | 1 448–1 490 | 929 | Proprietary |
| 44 | Kimi K2.6 | Moonshot AI | 1 466 | 1 450–1 483 | 1 501 | Modified MIT |
| 45 | Gemma 4 26B A4B | 1 466 | 1 428–1 504 | 250 | Apache 2.0 | |
| 46 | Grok 4.20 Multi-Agent | xAI | 1 466 | 1 451–1 480 | 2 181 | Proprietary |
| 47 | Qwen3.5 397B A17B | Alibaba Qwen | 1 464 | 1 451–1 478 | 2 534 | Apache 2.0 |
| 48 | Claude Sonnet 4.6 | Anthropic | 1 463 | 1 449–1 478 | 2 158 | Proprietary |
| 49 | Qwen3.8 27B | Alibaba Qwen | 1 462 | 1 430–1 494 | 354 | Apache 2.0 |
| 50 | GLM 4.6 | Z.ai | 1 461 | 1 438–1 484 | 694 | MIT |
| 51 | Hy3 | Tencent | 1 461 | 1 427–1 495 | 340 | Apache 2.0 |
| 52 | Grok 3 | xAI | 1 461 | 1 430–1 492 | 392 | Proprietary |
| 53 | Mistral Medium 3.1 | Mistral AI | 1 460 | 1 446–1 475 | 2 043 | Proprietary |
| 54 | Grok 4.6 в режиме high | xAI | 1 460 | 1 431–1 490 | 421 | Proprietary |
| 55 | Mistral Large 3 2512 | Mistral AI | 1 460 | 1 445–1 475 | 1 945 | Apache 2.0 |
| 56 | DeepSeek V4 Pro 0813 в режиме high | DeepSeek | 1 459 | 1 425–1 494 | 286 | MIT |
| 57 | Claude Opus 4.5 в режиме high | Anthropic | 1 459 | 1 437–1 480 | 825 | Proprietary |
| 58 | LongCat-Flash-Chat · chat | Meituan | 1 458 | 1 414–1 502 | 182 | MIT |
| 59 | Claude Sonnet 4.5 | Anthropic | 1 458 | 1 443–1 473 | 1 952 | Proprietary |
| 60 | ERNIE 5.0 · 0110 | Baidu | 1 458 | 1 439–1 477 | 1 043 | Proprietary |
| 61 | GLM 5 | Z.ai | 1 457 | 1 437–1 478 | 870 | MIT |
| 62 | Qwen3.6 Max Preview | Alibaba Qwen | 1 457 | 1 416–1 499 | 210 | Proprietary |
| 63 | Claude Sonnet 5 в режиме high | Anthropic | 1 457 | 1 438–1 476 | 1 202 | Proprietary |
| 64 | GPT-5.6 Luna в режиме xhigh | OpenAI | 1 456 | 1 436–1 476 | 963 | Proprietary |
| 65 | Grok 4.5 | xAI | 1 455 | 1 434–1 475 | 997 | Proprietary |
| 66 | Gemini 3.5 Flash Lite | 1 454 | 1 432–1 475 | 850 | Proprietary | |
| 67 | Kimi K2.5 · thinking | Moonshot AI | 1 453 | 1 440–1 467 | 2 529 | Modified MIT |
| 68 | Qwen3.6 Plus | Alibaba Qwen | 1 453 | 1 437–1 469 | 1 762 | Proprietary |
| 69 | GPT-5.2 в режиме high | OpenAI | 1 453 | 1 435–1 470 | 1 381 | Proprietary |
| 70 | Claude Sonnet 4.5 в режиме high | Anthropic | 1 452 | 1 437–1 467 | 1 937 | Proprietary |
| 71 | DeepSeek V4 Pro 0423 в режиме high | DeepSeek | 1 452 | 1 437–1 467 | 1 979 | MIT |
| 72 | Qwen3 Max | Alibaba Qwen | 1 452 | 1 423–1 480 | 431 | Proprietary |
| 73 | Inkling | Thinking Machines Lab | 1 451 | 1 429–1 473 | 833 | Apache 2.0 |
| 74 | Grok 4.1 · 4.1-thinking | xAI | 1 451 | 1 435–1 467 | 1 658 | Proprietary |
| 75 | Gemini 3 Flash Preview в режиме minimal | 1 450 | 1 437–1 463 | 2 723 | Proprietary | |
| 76 | Amazon Nova Experimental Chat · 11-10 | Amazon | 1 450 | 1 425–1 474 | 614 | Proprietary |
| 77 | Grok 4.20 · beta1 | xAI | 1 450 | 1 428–1 472 | 817 | Proprietary |
| 78 | Qwen3 VL 235B A22B Instruct | Alibaba Qwen | 1 449 | 1 409–1 490 | 255 | Apache 2.0 |
| 79 | MiMo-V2 Omni | Xiaomi | 1 448 | 1 426–1 471 | 771 | Proprietary |
| 80 | Mistral Medium 3.5 | Mistral AI | 1 448 | 1 418–1 479 | 435 | Modified MIT |
| 81 | DeepSeek V3.1 · v3.1 | DeepSeek | 1 448 | 1 414–1 483 | 286 | MIT |
| 82 | ChatGPT-4o (latest) | OpenAI | 1 448 | 1 431–1 465 | 1 386 | Proprietary |
| 83 | GLM 5V Turbo | Z.ai | 1 448 | 1 417–1 478 | 394 | Proprietary |
| 84 | GPT-5.1 в режиме high | OpenAI | 1 448 | 1 427–1 468 | 922 | Proprietary |
| 85 | MiMo-V2.5 | Xiaomi | 1 447 | 1 431–1 463 | 1 748 | MIT |
| 86 | Grok 4.1 · 4.1 | xAI | 1 446 | 1 431–1 462 | 1 743 | Proprietary |
| 87 | Qwen3 235B A22B Instruct 2507 | Alibaba Qwen | 1 445 | 1 431–1 459 | 2 112 | Apache 2.0 |
| 88 | MiniMax M3 | MiniMax | 1 444 | 1 428–1 460 | 1 782 | MiniMax Community License |
| 89 | Qwen3.5-122B-A10B | Alibaba Qwen | 1 443 | 1 422–1 464 | 898 | Apache 2.0 |
| 90 | GPT-5.2 Chat | OpenAI | 1 441 | 1 421–1 461 | 989 | Proprietary |
| 91 | Inkling Small | Thinking Machines Lab | 1 440 | 1 414–1 465 | 589 | Apache 2.0 |
| 92 | Gemma 4 31B | 1 439 | 1 403–1 474 | 261 | Apache 2.0 | |
| 93 | GPT-5.4 Mini в режиме high | OpenAI | 1 438 | 1 423–1 453 | 2 129 | Proprietary |
| 94 | DeepSeek V3.2 Exp · exp | DeepSeek | 1 437 | 1 397–1 477 | 224 | MIT |
| 95 | GLM 4.7 | Z.ai | 1 436 | 1 402–1 470 | 300 | MIT |
| 96 | GPT-5.2 | OpenAI | 1 436 | 1 422–1 449 | 2 644 | Proprietary |
| 97 | DeepSeek V4 Flash 0423 | DeepSeek | 1 436 | 1 420–1 451 | 1 913 | MIT |
| 98 | DeepSeek V4 Flash 0423 в режиме high | DeepSeek | 1 435 | 1 419–1 451 | 1 870 | MIT |
| 99 | Claude Opus 4.1 · 20250805-thinking-16k | Anthropic | 1 433 | 1 412–1 455 | 834 | Proprietary |
| 100 | MiMo-V2 Flash · flash (thinking) | Xiaomi | 1 433 | 1 402–1 464 | 338 | MIT |
| 101 | DeepSeek V3.2 · v3.2 | DeepSeek | 1 433 | 1 415–1 451 | 1 196 | MIT |
| 102 | Amazon Nova Experimental Chat · 10-20 | Amazon | 1 433 | 1 395–1 470 | 248 | Proprietary |
| 103 | Gemini 2.5 Flash · flash-preview-09-2025 | 1 432 | 1 407–1 457 | 576 | Proprietary | |
| 104 | o3 | OpenAI | 1 431 | 1 409–1 454 | 822 | Proprietary |
| 105 | GPT-5.1 | OpenAI | 1 431 | 1 411–1 452 | 891 | Proprietary |
| 106 | Gemini 2.5 Flash · flash | 1 431 | 1 418–1 444 | 2 540 | Proprietary | |
| 107 | Grok 4 Fast | xAI | 1 430 | 1 395–1 465 | 284 | Proprietary |
| 108 | Claude Opus 4.1 · 20250805 | Anthropic | 1 427 | 1 410–1 444 | 1 389 | Proprietary |
| 109 | Hy3 preview | Tencent | 1 427 | 1 387–1 466 | 236 | tencent-hunyuan-community |
| 110 | Kimi K2 Thinking | Moonshot AI | 1 425 | 1 409–1 441 | 1 580 | Modified MIT |
| 111 | GPT-5.5 · 5.5-instant | OpenAI | 1 425 | 1 405–1 445 | 1 015 | Proprietary |
| 112 | Gemini 3.1 Flash Lite Preview | 1 424 | 1 409–1 439 | 2 031 | Proprietary | |
| 113 | DeepSeek V3.2 · v3.2-thinking | DeepSeek | 1 423 | 1 404–1 442 | 1 013 | MIT |
| 114 | GLM 4.5 | Z.ai | 1 423 | 1 391–1 455 | 337 | MIT |
| 115 | MiMo-V2 Flash · flash (non-thinking) | Xiaomi | 1 421 | 1 404–1 439 | 1 299 | MIT |
| 116 | Step 3.5 Flash | StepFun | 1 421 | 1 406–1 436 | 1 957 | Apache 2.0 |
| 117 | Qwen3 30B A3B Instruct 2507 | Alibaba Qwen | 1 420 | 1 388–1 452 | 333 | Apache 2.0 |
| 118 | MiniMax M2.7 | MiniMax | 1 420 | 1 406–1 433 | 2 679 | Modified MIT |
| 119 | Grok 4 | xAI | 1 419 | 1 394–1 444 | 589 | Proprietary |
| 120 | GPT-4.5 Preview | OpenAI | 1 418 | 1 362–1 474 | 121 | Proprietary |
| 121 | Qwen3 Next 80B A3B Instruct | Alibaba Qwen | 1 417 | 1 386–1 448 | 376 | Apache 2.0 |
| 122 | R1 0528 | DeepSeek | 1 416 | 1 378–1 454 | 236 | MIT |
| 123 | Grok 4.1 Fast | xAI | 1 415 | 1 399–1 432 | 1 382 | Proprietary |
| 124 | DeepSeek V3.2 Exp · exp-thinking | DeepSeek | 1 414 | 1 368–1 460 | 145 | MIT |
| 125 | Qwen3.5-35B-A3B | Alibaba Qwen | 1 413 | 1 391–1 434 | 850 | Apache 2.0 |
| 126 | Grok 4.3 | xAI | 1 412 | 1 398–1 426 | 2 474 | Proprietary |
| 127 | LongCat-Flash-Chat · chat-2602-exp | Meituan | 1 411 | 1 390–1 432 | 952 | Proprietary |
| 128 | Qwen3.5-Flash | Alibaba Qwen | 1 411 | 1 396–1 425 | 2 176 | Proprietary |
| 129 | GPT-5 | OpenAI | 1 410 | 1 381–1 439 | 427 | Proprietary |
| 130 | GPT-5 в режиме high | OpenAI | 1 409 | 1 380–1 438 | 473 | Proprietary |
| 131 | Qwen3.5-27B | Alibaba Qwen | 1 407 | 1 386–1 429 | 815 | Apache 2.0 |
| 132 | Claude Haiku 4.5 | Anthropic | 1 407 | 1 395–1 419 | 3 651 | Proprietary |
| 133 | Kimi K2.5 · instant | Moonshot AI | 1 405 | 1 358–1 451 | 154 | Modified MIT |
| 134 | Nemotron 3 Super | NVIDIA | 1 404 | 1 368–1 440 | 267 | NVIDIA Open Model |
| 135 | DeepSeek V3.1 · v3.1-thinking | DeepSeek | 1 402 | 1 358–1 446 | 183 | MIT |
| 136 | GLM 4.5 Air | Z.ai | 1 401 | 1 374–1 428 | 511 | MIT |
| 137 | GPT-5.3 Chat | OpenAI | 1 400 | 1 380–1 420 | 1 009 | Proprietary |
| 138 | MiniMax M2.1 | MiniMax | 1 398 | 1 370–1 427 | 435 | MIT |
| 139 | GPT-5.4 Nano в режиме high | OpenAI | 1 398 | 1 382–1 413 | 1 991 | Proprietary |
| 140 | Nemotron 3 Nano 30B A3B | NVIDIA | 1 397 | 1 367–1 428 | 366 | NVIDIA Open Model |
| 141 | Qwen2.5 Max | Alibaba Qwen | 1 396 | 1 364–1 428 | 343 | Proprietary |
| 142 | Qwen3 Coder 480B A35B | Alibaba Qwen | 1 396 | 1 366–1 426 | 440 | Apache 2.0 |
| 143 | Hunyuan TurboS | Tencent | 1 391 | 1 343–1 439 | 150 | Proprietary |
| 144 | Gemini 2.0 Flash | 1 390 | 1 364–1 416 | 538 | Proprietary | |
| 145 | Gemini 2.5 Flash Lite · 09-2025-no-thinking | 1 390 | 1 369–1 411 | 825 | Proprietary | |
| 146 | DeepSeek V3 0324 | DeepSeek | 1 386 | 1 362–1 411 | 590 | MIT |
| 147 | GPT-5 Mini в режиме high | OpenAI | 1 386 | 1 355–1 416 | 407 | Proprietary |
| 148 | Gemini 2.5 Flash Lite · 06-17-thinking | 1 385 | 1 357–1 414 | 439 | Proprietary | |
| 149 | GPT-4.1 | OpenAI | 1 383 | 1 361–1 406 | 734 | Proprietary |
| 150 | Nova 2 Lite | Amazon | 1 382 | 1 345–1 419 | 258 | Proprietary |
| 151 | Kimi K2 0905 | Moonshot AI | 1 378 | 1 339–1 416 | 217 | Modified MIT |
| 152 | Nemotron 3.5 Lightning | NVIDIA | 1 375 | 1 339–1 412 | 265 | OpenMDW-1.1 |
| 153 | Trinity Large | Arcee AI | 1 375 | 1 353–1 396 | 861 | Apache 2.0 |
| 154 | Kimi K2 0711 | Moonshot AI | 1 374 | 1 345–1 404 | 404 | Modified MIT |
| 155 | gpt-oss-120b | OpenAI | 1 372 | 1 344–1 401 | 452 | Apache 2.0 |
| 156 | Claude Opus 4 · 20250514-thinking-16k | Anthropic | 1 371 | 1 344–1 398 | 518 | Proprietary |
| 157 | MiniMax M1 | MiniMax | 1 370 | 1 343–1 398 | 465 | Apache 2.0 |
| 158 | Mistral Medium 3 | Mistral AI | 1 369 | 1 343–1 396 | 496 | Proprietary |
| 159 | Gemma 3 27B | 1 369 | 1 345–1 393 | 652 | Gemma | |
| 160 | Qwen3 235B A22B · a22b-no-thinking | Alibaba Qwen | 1 368 | 1 342–1 394 | 527 | Apache 2.0 |
| 161 | Trinity Large Thinking | Arcee AI | 1 366 | 1 346–1 387 | 1 003 | Apache 2.0 |
| 162 | R1 | DeepSeek | 1 366 | 1 324–1 408 | 188 | MIT |
| 163 | Claude Sonnet 4 · 20250514 | Anthropic | 1 364 | 1 339–1 389 | 564 | Proprietary |
| 164 | o4 Mini | OpenAI | 1 363 | 1 339–1 388 | 612 | Proprietary |
| 165 | MiniMax M2.5 | MiniMax | 1 362 | 1 344–1 381 | 1 256 | Modified MIT |
| 166 | Qwen3 235B A22B · a22b | Alibaba Qwen | 1 361 | 1 330–1 392 | 361 | Apache 2.0 |
| 167 | Claude Opus 4 · 20250514 | Anthropic | 1 361 | 1 336–1 385 | 629 | Proprietary |
| 168 | Grok 3 Mini | xAI | 1 358 | 1 321–1 395 | 274 | Proprietary |
| 169 | GPT-4.1 Mini | OpenAI | 1 358 | 1 332–1 383 | 580 | Proprietary |
| 170 | Claude Sonnet 4 · 20250514-thinking-32k | Anthropic | 1 357 | 1 330–1 384 | 465 | Proprietary |
| 171 | Command A | Cohere | 1 353 | 1 331–1 375 | 769 | CC-BY-NC-4.0 |
| 172 | Qwen3 Next 80B A3B Thinking | Alibaba Qwen | 1 352 | 1 312–1 392 | 215 | Apache 2.0 |
| 173 | Qwen3 30B A3B | Alibaba Qwen | 1 351 | 1 321–1 381 | 351 | Apache 2.0 |
| 174 | Grok 3 Mini в режиме high | xAI | 1 347 | 1 304–1 391 | 169 | Proprietary |
| 175 | Gemini 2.0 Flash-Lite | 1 347 | 1 309–1 385 | 253 | Proprietary | |
| 176 | o1 · preview | OpenAI | 1 344 | 1 309–1 379 | 290 | Proprietary |
| 177 | o1 · 2024-12-17 | OpenAI | 1 344 | 1 307–1 381 | 252 | Proprietary |
| 178 | DeepSeek V3 | DeepSeek | 1 342 | 1 305–1 380 | 242 | DeepSeek |
| 179 | Mistral Small 3.2 24B | Mistral AI | 1 338 | 1 299–1 377 | 226 | Apache 2.0 |
| 180 | MiniMax M2 | MiniMax | 1 338 | 1 283–1 393 | 133 | Apache 2.0 |
| 181 | QwQ 32B | Alibaba Qwen | 1 337 | 1 305–1 370 | 315 | Apache 2.0 |
| 182 | o3 Mini High | OpenAI | 1 334 | 1 289–1 378 | 203 | Proprietary |
| 183 | GLM 4.7 Flash | Z.ai | 1 333 | 1 296–1 370 | 256 | MIT |
| 184 | Athene V2 Chat | Nexusflow | 1 332 | 1 295–1 370 | 228 | NexusFlow |
| 185 | Gemma 3n E4B | 1 331 | 1 298–1 364 | 326 | Gemma | |
| 186 | Qwen Max | Alibaba Qwen | 1 330 | 1 288–1 371 | 183 | Qwen |
| 187 | OLMo 3.1 32B Instruct | Ai2 | 1 328 | 1 293–1 364 | 323 | Apache 2.0 |
| 188 | Mistral Large · 2411 | Mistral AI | 1 325 | 1 284–1 365 | 204 | MRL |
| 189 | Grok 2 | xAI | 1 317 | 1 291–1 344 | 604 | Proprietary |
| 190 | o3 Mini | OpenAI | 1 312 | 1 289–1 335 | 705 | Proprietary |
| 191 | GLM-4-Plus | Z.ai | 1 308 | 1 271–1 344 | 286 | Proprietary |
| 192 | Gemini 1.5 Pro · advanced-0514 | 1 306 | 1 280–1 332 | 754 | Proprietary | |
| 193 | Yi-Lightning | 01.AI | 1 305 | 1 268–1 342 | 270 | Proprietary |
| 194 | Claude 3.5 Sonnet · 20241022 | Anthropic | 1 305 | 1 283–1 326 | 844 | Proprietary |
| 195 | GPT-4o (2024-05-13) | OpenAI | 1 304 | 1 283–1 325 | 1 359 | Proprietary |
| 196 | Claude 3.7 Sonnet · 20250219 | Anthropic | 1 303 | 1 276–1 329 | 545 | Proprietary |
| 197 | Gemini 1.5 Pro · 1.5-pro-002 | 1 301 | 1 273–1 330 | 459 | Proprietary | |
| 198 | GPT-4o-mini (2024-07-18) | OpenAI | 1 296 | 1 272–1 321 | 649 | Proprietary |
| 199 | Claude 3.5 Sonnet · 20240620 | Anthropic | 1 293 | 1 270–1 317 | 936 | Proprietary |
| 200 | o1-mini | OpenAI | 1 293 | 1 263–1 323 | 424 | Proprietary |
| 201 | DeepSeek V2.5 | DeepSeek | 1 289 | 1 250–1 327 | 222 | DeepSeek |
| 202 | Claude 3.7 Sonnet · 20250219-thinking-32k | Anthropic | 1 288 | 1 260–1 316 | 510 | Proprietary |
| 203 | Athene 70B | Nexusflow | 1 286 | 1 243–1 329 | 199 | CC-BY-NC-4.0 |
| 204 | Grok 2 Mini | xAI | 1 286 | 1 257–1 315 | 471 | Proprietary |
| 205 | Llama 4 Scout | Meta | 1 284 | 1 254–1 314 | 423 | Llama |
| 206 | GPT-4 Turbo Preview · 1106-preview | OpenAI | 1 282 | 1 261–1 304 | 1 734 | Proprietary |
| 207 | GPT-4 Turbo Preview · 0125-preview | OpenAI | 1 281 | 1 258–1 303 | 1 516 | Proprietary |
| 208 | Llama 3.3 70B Instruct | Meta | 1 280 | 1 255–1 305 | 570 | Llama-3.3 |
| 209 | Qwen2.5 72B Instruct | Alibaba Qwen | 1 280 | 1 247–1 312 | 340 | Qwen |
| 210 | Llama 3.1 405B Instruct · bf16 | Meta | 1 278 | 1 248–1 309 | 364 | Llama 3.1 Community |
| 211 | GPT-4 Turbo | OpenAI | 1 275 | 1 253–1 297 | 1 402 | Proprietary |
| 212 | Claude 3 Opus | Anthropic | 1 275 | 1 256–1 294 | 2 847 | Proprietary |
| 213 | Mistral Large 2407 | Mistral AI | 1 273 | 1 243–1 302 | 441 | Mistral Research |
| 214 | Gemini 1.5 Pro · 1.5-pro-001 | 1 272 | 1 249–1 295 | 1 076 | Proprietary | |
| 215 | Nova Pro 1.0 | Amazon | 1 271 | 1 233–1 309 | 222 | Proprietary |
| 216 | Llama 3.1 405B Instruct · fp8 | Meta | 1 271 | 1 243–1 298 | 545 | Llama 3.1 Community |
| 217 | Mistral Small 3.1 24B | Mistral AI | 1 270 | 1 241–1 299 | 454 | Apache 2.0 |
| 218 | Magistral Medium | Mistral AI | 1 268 | 1 221–1 314 | 183 | Proprietary |
| 219 | Claude 3.5 Haiku | Anthropic | 1 263 | 1 241–1 285 | 827 | Proprietary |
| 220 | GPT-4o (2024-08-06) | OpenAI | 1 262 | 1 234–1 291 | 457 | Proprietary |
| 221 | OLMo 3.1 32B Think | Ai2 | 1 261 | 1 217–1 305 | 211 | Apache 2.0 |
| 222 | Llama 3.1 70B Instruct | Meta | 1 261 | 1 232–1 289 | 501 | Llama 3.1 Community |
| 223 | Llama 4 Maverick | Meta | 1 259 | 1 233–1 284 | 593 | Llama 4 |
| 224 | Gemini 1.5 Flash · 002 | 1 258 | 1 223–1 292 | 295 | Proprietary | |
| 225 | Aya Expanse 32B | Cohere | 1 248 | 1 213–1 282 | 274 | CC-BY-NC-4.0 |
| 226 | Gemma 2 27B | 1 247 | 1 224–1 270 | 793 | Gemma license | |
| 227 | Gemini 1.5 Flash · 001 | 1 240 | 1 216–1 265 | 919 | Proprietary | |
| 228 | Nova Micro 1.0 | Amazon | 1 237 | 1 193–1 282 | 163 | Proprietary |
| 229 | Nova Lite 1.0 | Amazon | 1 237 | 1 193–1 282 | 170 | Proprietary |
| 230 | Gemini 1.5 Flash-8B | 1 233 | 1 200–1 267 | 331 | Proprietary | |
| 231 | Llama 3 70B Instruct | Meta | 1 232 | 1 212–1 252 | 2 499 | Llama 3 Community |
| 232 | Claude 3 Sonnet | Anthropic | 1 229 | 1 208–1 250 | 1 824 | Proprietary |
| 233 | Phi 4 | Microsoft | 1 224 | 1 184–1 264 | 205 | MIT |
| 234 | GPT-4 · 0314 | OpenAI | 1 219 | 1 193–1 245 | 839 | Proprietary |
| 235 | Mistral Large · 2402 | Mistral AI | 1 209 | 1 185–1 233 | 1 049 | Proprietary |
| 236 | Command R+ | Cohere | 1 208 | 1 185–1 231 | 1 226 | CC-BY-NC-4.0 |
| 237 | Mistral Small 3 | Mistral AI | 1 203 | 1 154–1 252 | 142 | Apache 2.0 |
| 238 | Nemotron-4 340B Instruct | NVIDIA | 1 196 | 1 162–1 230 | 357 | NVIDIA Open Model |
| 239 | Claude 3 Haiku | Anthropic | 1 194 | 1 174–1 215 | 1 772 | Proprietary |
| 240 | Gemma 2 9B | 1 190 | 1 163–1 216 | 591 | Gemma license | |
| 241 | DeepSeek Coder V2 | DeepSeek | 1 184 | 1 143–1 226 | 207 | DeepSeek License |
| 242 | Mistral Medium (2023) | Mistral AI | 1 179 | 1 152–1 207 | 721 | Proprietary |
| 243 | Llama 3.1 8B Instruct | Meta | 1 177 | 1 147–1 207 | 478 | Llama 3.1 Community |
| 244 | Reka Flash 21B · 20240226 | Reka AI | 1 174 | 1 142–1 205 | 430 | Proprietary |
| 245 | Qwen2 72B Instruct | Alibaba Qwen | 1 170 | 1 143–1 197 | 596 | Qianwen LICENSE |
| 246 | GPT-4 · 0613 | OpenAI | 1 170 | 1 148–1 192 | 1 545 | Proprietary |
| 247 | Mixtral 8x7B Instruct | Mistral AI | 1 166 | 1 143–1 189 | 1 252 | Apache 2.0 |
| 248 | Mixtral 8x22B Instruct | Mistral AI | 1 166 | 1 140–1 192 | 767 | Apache 2.0 |
| 249 | Reka Flash 21B · 20240226-online | Reka AI | 1 162 | 1 121–1 203 | 230 | Proprietary |
| 250 | Command R | Cohere | 1 162 | 1 137–1 187 | 843 | CC-BY-NC-4.0 |
| 251 | Qwen1.5 72B Chat | Alibaba Qwen | 1 158 | 1 133–1 184 | 858 | Qianwen LICENSE |
| 252 | Llama 3 8B Instruct | Meta | 1 158 | 1 137–1 180 | 1 624 | Llama 3 Community |
| 253 | Gemma 2 2B | 1 157 | 1 127–1 187 | 469 | Gemma license | |
| 254 | Yi-1.5 34B Chat | 01.AI | 1 156 | 1 124–1 188 | 399 | Apache-2.0 |
| 255 | Qwen1.5 110B Chat | Alibaba Qwen | 1 151 | 1 118–1 184 | 375 | Qianwen LICENSE |
| 256 | Gemini 1.0 Pro | 1 145 | 1 112–1 177 | 462 | Proprietary | |
| 257 | Phi-3 Small 8K Instruct | Microsoft | 1 134 | 1 098–1 170 | 271 | MIT |
| 258 | Snowflake Arctic Instruct | Snowflake | 1 131 | 1 101–1 162 | 527 | Apache 2.0 |
| 259 | GPT-3.5 Turbo · 0125 | OpenAI | 1 117 | 1 094–1 140 | 1 284 | Proprietary |
| 260 | Starling-LM 7B Beta | Berkeley Nest | 1 116 | 1 076–1 155 | 248 | Apache-2.0 |
| 261 | Qwen1.5 14B Chat | Alibaba Qwen | 1 116 | 1 077–1 154 | 279 | Qianwen LICENSE |
| 262 | Phi-3 Medium 4K Instruct | Microsoft | 1 110 | 1 076–1 144 | 350 | MIT |
| 263 | Qwen1.5 32B Chat | Alibaba Qwen | 1 100 | 1 065–1 136 | 313 | Qianwen LICENSE |
| 264 | DBRX Instruct | Databricks | 1 096 | 1 065–1 126 | 538 | DBRX LICENSE |
| 265 | OpenChat 3.5 | OpenChat | 1 092 | 1 048–1 135 | 211 | Apache-2.0 |
| 266 | Llama 2 70B Chat | Meta | 1 090 | 1 062–1 118 | 679 | Llama 2 Community |
| 267 | Vicuna 33B | LMSYS | 1 085 | 1 047–1 123 | 295 | Non-commercial |
| 268 | Starling-LM 7B Alpha | Berkeley Nest | 1 085 | 1 041–1 128 | 210 | CC-BY-NC-4.0 |
| 269 | Yi 34B Chat | 01.AI | 1 081 | 1 040–1 122 | 234 | Yi License |
| 270 | Phi-3 Mini 4K Instruct | Microsoft | 1 076 | 1 039–1 112 | 307 | MIT |
| 271 | GPT-3.5 Turbo · 1106 | OpenAI | 1 066 | 1 027–1 105 | 310 | Proprietary |
| 272 | Gemma 1.1 7B | 1 065 | 1 032–1 098 | 401 | Gemma license | |
| 273 | Llama 2 13B Chat | Meta | 1 044 | 1 004–1 084 | 262 | Llama 2 Community |
| 274 | Phi-3 Mini 128K Instruct | Microsoft | 1 039 | 1 003–1 074 | 372 | MIT |
| 275 | Mistral 7B Instruct · instruct-v0.2 | Mistral AI | 1 036 | 1 004–1 069 | 473 | Apache-2.0 |
| 276 | StripedHyena Nous 7B | Together AI | 1 032 | 982–1 081 | 169 | Apache 2.0 |
| 277 | Zephyr 7B Beta | Hugging Face H4 | 1 030 | 981–1 080 | 153 | MIT |
| 278 | Gemma 7B | 1 025 | 977–1 073 | 199 | Gemma license | |
| 279 | Vicuna 13B | LMSYS | 1 020 | 974–1 065 | 198 | Llama 2 Community |
| 280 | Llama 2 7B Chat | Meta | 969 | 926–1 013 | 257 | Llama 2 Community |
| 281 | Mistral 7B Instruct · instruct | Mistral AI | 945 | 897–994 | 175 | Apache 2.0 |
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.