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 «без ничьих», 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 | 1 527 | 1 516–1 538 | 4 305 | Proprietary |
| 2 | Claude Opus 5 в режиме max | Anthropic | 1 522 | 1 514–1 529 | 15 658 | Proprietary |
| 3 | Claude Opus 5 в режиме high | Anthropic | 1 521 | 1 516–1 527 | 32 458 | Proprietary |
| 4 | Claude Opus 4.6 в режиме high | Anthropic | 1 519 | 1 514–1 523 | 54 935 | Proprietary |
| 5 | Claude Opus 4.6 | Anthropic | 1 513 | 1 509–1 518 | 57 679 | Proprietary |
| 6 | Gemini 3.8 Flash в режиме high | 1 510 | 1 498–1 522 | 3 735 | Proprietary | |
| 7 | Claude Fable 5 | Anthropic | 1 506 | 1 500–1 512 | 22 651 | Proprietary |
| 8 | Muse Spark 1.3 в режиме max | Meta | 1 503 | 1 491–1 515 | 3 495 | Proprietary |
| 9 | Muse Spark 1.2 в режиме xhigh | Meta | 1 502 | 1 488–1 517 | 2 391 | Proprietary |
| 10 | Claude Opus 4.7 в режиме high | Anthropic | 1 502 | 1 497–1 507 | 45 741 | Proprietary |
| 11 | Gemini 3.7 Flash в режиме high | 1 501 | 1 490–1 512 | 4 324 | Proprietary | |
| 12 | Claude Opus 4.7 | Anthropic | 1 492 | 1 487–1 497 | 46 678 | Proprietary |
| 13 | Gemini 3 Pro | 1 491 | 1 486–1 497 | 29 202 | Proprietary | |
| 14 | Gemini 3.5 Flash в режиме high | 1 491 | 1 485–1 496 | 28 967 | Proprietary | |
| 15 | Qwen3.8 Max (0902) | Alibaba Qwen | 1 490 | 1 482–1 497 | 12 384 | Proprietary |
| 16 | Gemini 3.1 Pro Preview | 1 490 | 1 485–1 494 | 80 424 | Proprietary | |
| 17 | Muse Spark 1.1 | Meta | 1 488 | 1 482–1 495 | 20 551 | Proprietary |
| 18 | Gemini 3.6 Flash в режиме high | 1 484 | 1 477–1 490 | 19 732 | Proprietary | |
| 19 | Gemini 3.5 Flash в режиме medium | 1 483 | 1 477–1 489 | 27 560 | Proprietary | |
| 20 | GLM 5.3 в режиме max | Z.ai | 1 481 | 1 473–1 490 | 8 205 | MIT |
| 21 | Qwen3.5 Max | Alibaba Qwen | 1 479 | 1 473–1 486 | 15 738 | Proprietary |
| 22 | Muse Spark | Meta | 1 478 | 1 471–1 486 | 10 287 | Proprietary |
| 23 | Qwen3.7 Max | Alibaba Qwen | 1 478 | 1 465–1 491 | 2 846 | Proprietary |
| 24 | Kimi K3 в режиме max | Moonshot AI | 1 478 | 1 471–1 485 | 15 621 | Kimi K3 license |
| 25 | GLM 5.3 Flash | Z.ai | 1 477 | 1 468–1 485 | 7 404 | MIT |
| 26 | GPT-5.5 в режиме high | OpenAI | 1 476 | 1 471–1 481 | 49 241 | Proprietary |
| 27 | Gemini 3 Flash Preview | 1 475 | 1 469–1 481 | 21 457 | Proprietary | |
| 28 | GPT-5.4 в режиме high | OpenAI | 1 474 | 1 469–1 479 | 46 309 | Proprietary |
| 29 | ERNIE 5.1 | Baidu | 1 472 | 1 467–1 478 | 28 139 | Proprietary |
| 30 | GLM 5.2 в режиме max | Z.ai | 1 471 | 1 465–1 476 | 27 350 | MIT |
| 31 | GPT-5.5 · 5.5 | OpenAI | 1 469 | 1 464–1 474 | 50 520 | Proprietary |
| 32 | MiMo-V2.5-Pro | Xiaomi | 1 467 | 1 462–1 472 | 46 360 | MIT |
| 33 | GLM 5.1 | Z.ai | 1 465 | 1 460–1 470 | 36 606 | MIT |
| 34 | Claude Opus 4.8 в режиме high | Anthropic | 1 463 | 1 458–1 469 | 39 441 | Proprietary |
| 35 | Gemini 2.5 Pro | 1 461 | 1 457–1 464 | 89 403 | Proprietary | |
| 36 | Claude Sonnet 4.6 | Anthropic | 1 459 | 1 454–1 464 | 50 083 | Proprietary |
| 37 | GPT-5.6 Sol в режиме xhigh | OpenAI | 1 455 | 1 449–1 462 | 20 446 | Proprietary |
| 38 | Kimi K2.6 | Moonshot AI | 1 455 | 1 449–1 461 | 27 761 | Modified MIT |
| 39 | Qwen3.7 Plus | Alibaba Qwen | 1 454 | 1 448–1 459 | 28 921 | Proprietary |
| 40 | Claude Opus 4.8 | Anthropic | 1 453 | 1 448–1 458 | 40 185 | Proprietary |
| 41 | GPT-5.4 | OpenAI | 1 451 | 1 447–1 456 | 48 802 | Proprietary |
| 42 | Grok 4.20 · beta-0309-reasoning | xAI | 1 449 | 1 444–1 454 | 47 344 | Proprietary |
| 43 | Grok 4.5 | xAI | 1 449 | 1 443–1 455 | 22 597 | Proprietary |
| 44 | DeepSeek V4 Pro 0423 | DeepSeek | 1 449 | 1 444–1 454 | 40 992 | MIT |
| 45 | Claude Opus 4.5 | Anthropic | 1 449 | 1 444–1 453 | 51 974 | Proprietary |
| 46 | Amazon Nova Experimental Chat · 26-02-10 | Amazon | 1 448 | 1 434–1 462 | 2 317 | Proprietary |
| 47 | Grok 4.20 Multi-Agent | xAI | 1 447 | 1 442–1 452 | 45 924 | Proprietary |
| 48 | Seed 2.0 Pro | ByteDance Seed | 1 446 | 1 442–1 451 | 56 214 | Proprietary |
| 49 | Claude Opus 4.5 в режиме high | Anthropic | 1 445 | 1 440–1 450 | 26 195 | Proprietary |
| 50 | GLM 5 | Z.ai | 1 444 | 1 438–1 450 | 19 909 | MIT |
| 51 | GPT-5.6 Terra в режиме xhigh | OpenAI | 1 443 | 1 437–1 450 | 21 004 | Proprietary |
| 52 | ERNIE 5.0 · 0110 | Baidu | 1 443 | 1 438–1 449 | 24 824 | Proprietary |
| 53 | Kimi K2.5 · thinking | Moonshot AI | 1 443 | 1 438–1 447 | 51 275 | Modified MIT |
| 54 | Qwen3.6 Max Preview | Alibaba Qwen | 1 441 | 1 430–1 452 | 3 902 | Proprietary |
| 55 | GPT-6 Astra в режиме max | OpenAI | 1 441 | 1 426–1 456 | 2 092 | Proprietary |
| 56 | DeepSeek V4 Pro 0423 в режиме high | DeepSeek | 1 440 | 1 435–1 446 | 39 102 | MIT |
| 57 | ERNIE 5.0 · preview-1203 | Baidu | 1 440 | 1 431–1 450 | 6 772 | Proprietary |
| 58 | Grok 4.20 · beta1 | xAI | 1 440 | 1 434–1 446 | 20 167 | Proprietary |
| 59 | Gemma 4 31B | 1 440 | 1 429–1 451 | 4 050 | Apache 2.0 | |
| 60 | Nemotron 3 Ultra | NVIDIA | 1 439 | 1 431–1 448 | 8 088 | OpenMDW-1.1 |
| 61 | DeepSeek V4 Pro 0813 в режиме high | DeepSeek | 1 439 | 1 430–1 448 | 6 590 | MIT |
| 62 | Gemini 3 Flash Preview в режиме minimal | 1 439 | 1 435–1 443 | 63 635 | Proprietary | |
| 63 | GPT-5.1 в режиме high | OpenAI | 1 439 | 1 434–1 444 | 29 030 | Proprietary |
| 64 | Claude Sonnet 5 в режиме high | Anthropic | 1 438 | 1 432–1 444 | 26 518 | Proprietary |
| 65 | GLM 4.6 | Z.ai | 1 438 | 1 432–1 443 | 24 587 | MIT |
| 66 | Qwen3 Max · preview | Alibaba Qwen | 1 436 | 1 430–1 442 | 19 316 | Proprietary |
| 67 | Hy3 | Tencent | 1 435 | 1 425–1 444 | 6 024 | Apache 2.0 |
| 68 | Qwen3.8 27B | Alibaba Qwen | 1 434 | 1 426–1 443 | 7 884 | Apache 2.0 |
| 69 | Qwen3.5 397B A17B | Alibaba Qwen | 1 434 | 1 430–1 438 | 56 363 | Apache 2.0 |
| 70 | Claude Sonnet 4.5 | Anthropic | 1 434 | 1 430–1 437 | 58 217 | Proprietary |
| 71 | Inkling | Thinking Machines Lab | 1 434 | 1 427–1 440 | 19 133 | Apache 2.0 |
| 72 | GPT-5.2 Chat | OpenAI | 1 433 | 1 428–1 438 | 25 958 | Proprietary |
| 73 | GLM 4.7 | Z.ai | 1 432 | 1 423–1 440 | 8 243 | MIT |
| 74 | Grok 4.1 · 4.1-thinking | xAI | 1 432 | 1 427–1 436 | 47 334 | Proprietary |
| 75 | Grok 4.1 · 4.1 | xAI | 1 431 | 1 427–1 435 | 48 015 | Proprietary |
| 76 | MiMo-V2 Pro | Xiaomi | 1 431 | 1 424–1 437 | 18 440 | Proprietary |
| 77 | Qwen3.6 Plus | Alibaba Qwen | 1 431 | 1 425–1 436 | 33 993 | Proprietary |
| 78 | GLM 5V Turbo | Z.ai | 1 430 | 1 421–1 439 | 6 894 | Proprietary |
| 79 | Gemini 3.5 Flash Lite | 1 430 | 1 423–1 436 | 19 705 | Proprietary | |
| 80 | Gemma 4 26B A4B | 1 429 | 1 418–1 440 | 3 979 | Apache 2.0 | |
| 81 | Claude Sonnet 4.5 в режиме high | Anthropic | 1 427 | 1 423–1 430 | 59 541 | Proprietary |
| 82 | MiniMax M3 | MiniMax | 1 426 | 1 421–1 432 | 36 612 | MiniMax Community License |
| 83 | ERNIE 5.0 · preview-1022 | Baidu | 1 425 | 1 412–1 437 | 3 211 | Proprietary |
| 84 | DeepSeek V4 Flash 0423 | DeepSeek | 1 424 | 1 419–1 429 | 36 924 | MIT |
| 85 | GLM 4.5 | Z.ai | 1 423 | 1 416–1 429 | 16 908 | MIT |
| 86 | Grok 4.6 в режиме high | xAI | 1 422 | 1 414–1 429 | 11 519 | Proprietary |
| 87 | ChatGPT-4o (latest) | OpenAI | 1 422 | 1 418–1 426 | 57 567 | Proprietary |
| 88 | GPT-5.6 Luna в режиме xhigh | OpenAI | 1 421 | 1 415–1 427 | 21 447 | Proprietary |
| 89 | R1 0528 | DeepSeek | 1 419 | 1 412–1 427 | 13 037 | MIT |
| 90 | Mistral Large 3 2512 | Mistral AI | 1 419 | 1 415–1 423 | 50 247 | Apache 2.0 |
| 91 | Grok 3 | xAI | 1 419 | 1 413–1 425 | 22 975 | Proprietary |
| 92 | MiMo-V2.5 | Xiaomi | 1 417 | 1 411–1 423 | 34 068 | MIT |
| 93 | LongCat-Flash-Chat · chat-2602-exp | Meituan | 1 416 | 1 410–1 422 | 21 064 | Proprietary |
| 94 | Mistral Medium 3.1 | Mistral AI | 1 416 | 1 412–1 419 | 66 650 | Proprietary |
| 95 | DeepSeek V3.2 · v3.2 | DeepSeek | 1 415 | 1 410–1 420 | 32 878 | MIT |
| 96 | DeepSeek V3.2 Exp · exp-thinking | DeepSeek | 1 415 | 1 405–1 424 | 6 339 | MIT |
| 97 | DeepSeek V3.2 Exp · exp | DeepSeek | 1 414 | 1 405–1 423 | 8 269 | MIT |
| 98 | DeepSeek V4 Flash 0423 в режиме high | DeepSeek | 1 413 | 1 408–1 418 | 36 957 | MIT |
| 99 | LongCat-Flash-Chat · chat | Meituan | 1 412 | 1 403–1 421 | 7 899 | MIT |
| 100 | GPT-5.1 | OpenAI | 1 412 | 1 407–1 417 | 30 668 | Proprietary |
| 101 | GPT-4.5 Preview | OpenAI | 1 411 | 1 402–1 419 | 10 031 | Proprietary |
| 102 | Amazon Nova Experimental Chat · 12-10 | Amazon | 1 411 | 1 397–1 425 | 2 505 | Proprietary |
| 103 | MiMo-V2 Omni | Xiaomi | 1 411 | 1 403–1 418 | 14 601 | Proprietary |
| 104 | Mistral Medium 3.5 | Mistral AI | 1 410 | 1 402–1 419 | 8 285 | Modified MIT |
| 105 | Kimi K2.5 · instant | Moonshot AI | 1 409 | 1 400–1 419 | 5 516 | Modified MIT |
| 106 | GPT-5.5 · 5.5-instant | OpenAI | 1 409 | 1 402–1 415 | 20 237 | Proprietary |
| 107 | Qwen3 235B A22B Instruct 2507 | Alibaba Qwen | 1 408 | 1 405–1 412 | 68 092 | Apache 2.0 |
| 108 | Qwen3 VL 235B A22B Instruct | Alibaba Qwen | 1 408 | 1 399–1 417 | 7 930 | Apache 2.0 |
| 109 | DeepSeek V3.1 Terminus · terminus-thinking | DeepSeek | 1 408 | 1 393–1 422 | 2 420 | MIT |
| 110 | DeepSeek V3.2 · v3.2-thinking | DeepSeek | 1 407 | 1 402–1 412 | 28 378 | MIT |
| 111 | Claude Opus 4.1 · 20250805-thinking-16k | Anthropic | 1 407 | 1 402–1 411 | 35 104 | Proprietary |
| 112 | DeepSeek V3.1 · v3.1 | DeepSeek | 1 406 | 1 397–1 414 | 10 425 | MIT |
| 113 | Qwen3 Next 80B A3B Instruct | Alibaba Qwen | 1 405 | 1 398–1 411 | 15 907 | Apache 2.0 |
| 114 | Claude Opus 4.1 · 20250805 | Anthropic | 1 405 | 1 401–1 409 | 54 723 | Proprietary |
| 115 | Gemini 2.5 Flash · flash | 1 405 | 1 401–1 408 | 88 133 | Proprietary | |
| 116 | Qwen3.5-122B-A10B | Alibaba Qwen | 1 404 | 1 398–1 410 | 20 738 | Apache 2.0 |
| 117 | GPT-5.2 в режиме high | OpenAI | 1 403 | 1 399–1 408 | 35 310 | Proprietary |
| 118 | DeepSeek V3.1 · v3.1-thinking | DeepSeek | 1 403 | 1 394–1 412 | 8 153 | MIT |
| 119 | Amazon Nova Experimental Chat · 11-10 | Amazon | 1 403 | 1 397–1 409 | 17 547 | Proprietary |
| 120 | DeepSeek V3.1 Terminus · terminus | DeepSeek | 1 402 | 1 388–1 415 | 2 583 | MIT |
| 121 | Qwen3 235B A22B Thinking 2507 | Alibaba Qwen | 1 402 | 1 392–1 411 | 6 198 | Apache 2.0 |
| 122 | Gemini 3.1 Flash Lite Preview | 1 401 | 1 396–1 406 | 45 056 | Proprietary | |
| 123 | Kimi K2 Thinking | Moonshot AI | 1 400 | 1 396–1 405 | 43 810 | Modified MIT |
| 124 | GPT-5.2 | OpenAI | 1 398 | 1 394–1 403 | 59 203 | Proprietary |
| 125 | Qwen3 Max · 2025-09-23 | Alibaba Qwen | 1 398 | 1 389–1 407 | 6 233 | Proprietary |
| 126 | GPT-5.4 Mini в режиме high | OpenAI | 1 398 | 1 393–1 403 | 45 551 | Proprietary |
| 127 | MiMo-V2 Flash · flash (non-thinking) | Xiaomi | 1 397 | 1 392–1 401 | 33 283 | MIT |
| 128 | Inkling Small | Thinking Machines Lab | 1 396 | 1 389–1 404 | 13 900 | Apache 2.0 |
| 129 | Grok 4 | xAI | 1 396 | 1 390–1 401 | 28 363 | Proprietary |
| 130 | o3 | OpenAI | 1 393 | 1 388–1 398 | 42 657 | Proprietary |
| 131 | Grok 4.1 Fast | xAI | 1 392 | 1 387–1 396 | 40 077 | Proprietary |
| 132 | Qwen3.5-27B | Alibaba Qwen | 1 390 | 1 384–1 396 | 19 888 | Apache 2.0 |
| 133 | Grok 4 Fast · chat | xAI | 1 390 | 1 379–1 401 | 4 542 | Proprietary |
| 134 | Gemini 2.5 Flash · flash-preview-09-2025 | 1 389 | 1 384–1 395 | 22 994 | Proprietary | |
| 135 | Hy3 preview | Tencent | 1 389 | 1 379–1 398 | 4 979 | tencent-hunyuan-community |
| 136 | Hunyuan Vision 1.5 | Tencent | 1 388 | 1 371–1 406 | 1 557 | Proprietary |
| 137 | MiniMax M2.7 | MiniMax | 1 388 | 1 383–1 392 | 52 396 | Modified MIT |
| 138 | GPT-5 в режиме high | OpenAI | 1 387 | 1 381–1 393 | 23 206 | Proprietary |
| 139 | Step 3.5 Flash | StepFun | 1 385 | 1 380–1 390 | 41 610 | Apache 2.0 |
| 140 | GPT-5 | OpenAI | 1 385 | 1 379–1 390 | 22 224 | Proprietary |
| 141 | Hunyuan T1 | Tencent | 1 380 | 1 368–1 393 | 3 248 | Proprietary |
| 142 | Qwen3 VL 235B A22B Thinking | Alibaba Qwen | 1 379 | 1 370–1 389 | 5 477 | Apache 2.0 |
| 143 | Grok 4 Fast · reasoning | xAI | 1 379 | 1 372–1 386 | 12 694 | Proprietary |
| 144 | Grok 4.3 | xAI | 1 379 | 1 374–1 383 | 51 145 | Proprietary |
| 145 | Qwen3.5-Flash | Alibaba Qwen | 1 377 | 1 372–1 382 | 42 950 | Proprietary |
| 146 | Claude Haiku 4.5 | Anthropic | 1 376 | 1 373–1 380 | 96 498 | Proprietary |
| 147 | Amazon Nova Experimental Chat · 26-01-10 | Amazon | 1 375 | 1 361–1 390 | 2 294 | Proprietary |
| 148 | Qwen3.5-35B-A3B | Alibaba Qwen | 1 374 | 1 369–1 380 | 21 196 | Apache 2.0 |
| 149 | Amazon Nova Experimental Chat · 10-20 | Amazon | 1 374 | 1 366–1 383 | 8 082 | Proprietary |
| 150 | Qwen3 235B A22B · a22b-no-thinking | Alibaba Qwen | 1 371 | 1 365–1 377 | 26 597 | Apache 2.0 |
| 151 | MiMo-V2 Flash · flash (thinking) | Xiaomi | 1 370 | 1 361–1 379 | 7 484 | MIT |
| 152 | Muse Glimmer 30B | Meta | 1 369 | 1 355–1 383 | 2 711 | Apache-2.0 |
| 153 | MiniMax M2.1 | MiniMax | 1 367 | 1 360–1 375 | 11 714 | MIT |
| 154 | GPT-5.3 Chat | OpenAI | 1 366 | 1 361–1 372 | 25 125 | Proprietary |
| 155 | Qwen3 30B A3B Instruct 2507 | Alibaba Qwen | 1 358 | 1 351–1 364 | 16 321 | Apache 2.0 |
| 156 | GLM 4.5 Air | Z.ai | 1 357 | 1 351–1 363 | 21 330 | MIT |
| 157 | GPT-4.1 | OpenAI | 1 356 | 1 351–1 361 | 35 457 | Proprietary |
| 158 | Kimi K2 0905 | Moonshot AI | 1 351 | 1 343–1 360 | 8 409 | Modified MIT |
| 159 | Gemini 2.5 Flash Lite · 09-2025-no-thinking | 1 350 | 1 345–1 355 | 32 445 | Proprietary | |
| 160 | Solar Pro 4 | Upstage | 1 348 | 1 331–1 365 | 1 797 | Proprietary |
| 161 | Hunyuan TurboS · 20250416 | Tencent | 1 348 | 1 339–1 357 | 7 407 | Proprietary |
| 162 | GLM 4.6V | Z.ai | 1 347 | 1 332–1 363 | 1 972 | MIT |
| 163 | Nemotron 3 Super | NVIDIA | 1 347 | 1 337–1 357 | 5 135 | NVIDIA Open Model |
| 164 | Claude Opus 4 · 20250514-thinking-16k | Anthropic | 1 346 | 1 341–1 352 | 26 151 | Proprietary |
| 165 | GPT-5.4 Nano в режиме high | OpenAI | 1 346 | 1 341–1 351 | 44 779 | Proprietary |
| 166 | DeepSeek V3 0324 | DeepSeek | 1 346 | 1 340–1 351 | 31 347 | MIT |
| 167 | R1 | DeepSeek | 1 346 | 1 339–1 353 | 12 504 | MIT |
| 168 | GPT-5 Mini в режиме high | OpenAI | 1 342 | 1 336–1 348 | 19 354 | Proprietary |
| 169 | Mistral Medium 3 | Mistral AI | 1 340 | 1 334–1 347 | 22 704 | Proprietary |
| 170 | Kimi K2 0711 | Moonshot AI | 1 338 | 1 332–1 344 | 19 541 | Modified MIT |
| 171 | Qwen2.5 Max | Alibaba Qwen | 1 337 | 1 331–1 343 | 21 825 | Proprietary |
| 172 | Gemini 2.5 Flash Lite · 06-17-thinking | 1 336 | 1 330–1 342 | 22 685 | Proprietary | |
| 173 | Grok 3 Mini в режиме high | xAI | 1 335 | 1 327–1 342 | 11 080 | Proprietary |
| 174 | Qwen3 Next 80B A3B Thinking | Alibaba Qwen | 1 334 | 1 326–1 342 | 9 525 | Apache 2.0 |
| 175 | o1 · 2024-12-17 | OpenAI | 1 334 | 1 327–1 340 | 19 214 | Proprietary |
| 176 | Qwen3 235B A22B · a22b | Alibaba Qwen | 1 332 | 1 326–1 339 | 17 758 | Apache 2.0 |
| 177 | Granite 4.2 30B | IBM Granite | 1 332 | 1 317–1 346 | 2 458 | Apache 2.0 |
| 178 | Claude Opus 4 · 20250514 | Anthropic | 1 331 | 1 326–1 337 | 31 015 | Proprietary |
| 179 | gpt-oss-120b | OpenAI | 1 331 | 1 325–1 337 | 21 306 | Apache 2.0 |
| 180 | Grok 3 Mini | xAI | 1 329 | 1 322–1 336 | 14 677 | Proprietary |
| 181 | Amazon Nova Experimental Chat · 10-09 | Amazon | 1 328 | 1 313–1 344 | 1 927 | Proprietary |
| 182 | Nova 2 Lite | Amazon | 1 327 | 1 318–1 335 | 8 681 | Proprietary |
| 183 | MiniMax M2.5 | MiniMax | 1 326 | 1 321–1 331 | 30 850 | Modified MIT |
| 184 | Ling-flash-2.0 | inclusionAI | 1 324 | 1 313–1 334 | 4 837 | MIT |
| 185 | Gemma 3 27B | 1 321 | 1 316–1 326 | 32 532 | Gemma | |
| 186 | Qwen3 Coder 480B A35B | Alibaba Qwen | 1 318 | 1 312–1 325 | 17 916 | Apache 2.0 |
| 187 | Gemini 2.0 Flash | 1 317 | 1 312–1 323 | 29 828 | Proprietary | |
| 188 | INTELLECT-3 | Prime Intellect | 1 317 | 1 305–1 328 | 3 758 | MIT |
| 189 | Mercury 2 | Inception Labs | 1 316 | 1 300–1 331 | 2 118 | Proprietary |
| 190 | o4 Mini | OpenAI | 1 315 | 1 309–1 320 | 31 552 | Proprietary |
| 191 | o1 · preview | OpenAI | 1 314 | 1 307–1 321 | 21 521 | Proprietary |
| 192 | Step 3 | StepFun | 1 311 | 1 301–1 322 | 4 445 | Apache 2.0 |
| 193 | Claude Sonnet 4 · 20250514-thinking-32k | Anthropic | 1 310 | 1 304–1 316 | 24 897 | Proprietary |
| 194 | GLM 4.7 Flash | Z.ai | 1 309 | 1 301–1 318 | 7 973 | MIT |
| 195 | Trinity Large Thinking | Arcee AI | 1 306 | 1 299–1 312 | 22 508 | Apache 2.0 |
| 196 | Nemotron 3 Nano 30B A3B | NVIDIA | 1 304 | 1 296–1 312 | 10 848 | NVIDIA Open Model |
| 197 | MiniMax M2 | MiniMax | 1 299 | 1 288–1 309 | 4 914 | Apache 2.0 |
| 198 | MiniMax M1 | MiniMax | 1 298 | 1 292–1 304 | 23 914 | Apache 2.0 |
| 199 | Trinity Large | Arcee AI | 1 297 | 1 291–1 303 | 22 593 | Apache 2.0 |
| 200 | Qwen3 32B | Alibaba Qwen | 1 296 | 1 282–1 310 | 2 544 | Apache 2.0 |
| 201 | GPT-4.1 Mini | OpenAI | 1 296 | 1 290–1 302 | 27 420 | Proprietary |
| 202 | Claude Sonnet 4 · 20250514 | Anthropic | 1 295 | 1 289–1 301 | 28 475 | Proprietary |
| 203 | Llama 3.3 Nemotron Super 49B v1.5 | NVIDIA | 1 293 | 1 278–1 307 | 2 261 | Nvidia Open |
| 204 | Mistral Small 3.2 24B | Mistral AI | 1 292 | 1 285–1 300 | 12 145 | Apache 2.0 |
| 205 | o3 Mini High | OpenAI | 1 291 | 1 284–1 299 | 12 679 | Proprietary |
| 206 | Gemma 3 12B | 1 289 | 1 275–1 303 | 2 526 | Gemma | |
| 207 | DeepSeek V3 | DeepSeek | 1 289 | 1 281–1 296 | 14 106 | DeepSeek |
| 208 | Step-1o Turbo | StepFun | 1 288 | 1 279–1 298 | 6 333 | Proprietary |
| 209 | Nemotron 3.5 Lightning | NVIDIA | 1 284 | 1 274–1 294 | 6 305 | OpenMDW-1.1 |
| 210 | GLM-4-Plus · plus-0111 | Z.ai | 1 284 | 1 271–1 296 | 3 826 | Proprietary |
| 211 | Ring-flash-2.0 | inclusionAI | 1 283 | 1 273–1 294 | 4 909 | MIT |
| 212 | GLM 4.5V | Z.ai | 1 283 | 1 271–1 295 | 3 361 | MIT |
| 213 | Command A | Cohere | 1 281 | 1 276–1 286 | 38 271 | CC-BY-NC-4.0 |
| 214 | Gemini 2.0 Flash-Lite | 1 281 | 1 274–1 287 | 16 266 | Proprietary | |
| 215 | Qwen-Plus | Alibaba Qwen | 1 280 | 1 267–1 293 | 3 765 | Proprietary |
| 216 | QwQ 32B | Alibaba Qwen | 1 279 | 1 273–1 286 | 17 057 | Apache 2.0 |
| 217 | Step-2 16k | StepFun | 1 272 | 1 258–1 286 | 3 048 | Proprietary |
| 218 | Granite 4.2 8B | IBM Granite | 1 269 | 1 253–1 285 | 2 371 | Apache 2.0 |
| 219 | GPT-5 Nano в режиме high | OpenAI | 1 268 | 1 258–1 277 | 5 885 | Proprietary |
| 220 | Gemini 1.5 Pro · 1.5-pro-002 | 1 266 | 1 261–1 271 | 35 980 | Proprietary | |
| 221 | o3 Mini | OpenAI | 1 265 | 1 260–1 270 | 39 558 | Proprietary |
| 222 | Hunyuan TurboS · 20250226 | Tencent | 1 265 | 1 246–1 284 | 1 373 | Proprietary |
| 223 | Llama 3.1 Nemotron Ultra 253B v1 | NVIDIA | 1 263 | 1 246–1 281 | 1 699 | Nvidia Open Model |
| 224 | o1-mini | OpenAI | 1 262 | 1 257–1 268 | 34 410 | Proprietary |
| 225 | Claude 3.7 Sonnet · 20250219-thinking-32k | Anthropic | 1 261 | 1 255–1 267 | 28 135 | Proprietary |
| 226 | Qwen3 30B A3B | Alibaba Qwen | 1 260 | 1 253–1 267 | 18 057 | Apache 2.0 |
| 227 | Llama 3.3 Nemotron Super 49B v1 | NVIDIA | 1 252 | 1 233–1 271 | 1 430 | Nvidia |
| 228 | OLMo 3.1 32B Instruct | Ai2 | 1 251 | 1 241–1 260 | 8 186 | Apache 2.0 |
| 229 | Hunyuan Turbo | Tencent | 1 248 | 1 230–1 267 | 1 358 | Proprietary |
| 230 | Granite 4.2 3B | IBM Granite | 1 246 | 1 230–1 262 | 2 417 | Apache 2.0 |
| 231 | Gemma 3n E4B | 1 245 | 1 238–1 252 | 15 675 | Gemma | |
| 232 | Grok 2 | xAI | 1 244 | 1 238–1 249 | 40 763 | Proprietary |
| 233 | Yi-Lightning | 01.AI | 1 241 | 1 234–1 249 | 17 462 | Proprietary |
| 234 | Claude 3.7 Sonnet · 20250219 | Anthropic | 1 241 | 1 235–1 246 | 31 167 | Proprietary |
| 235 | OLMo 3 32B Think | Ai2 | 1 238 | 1 226–1 251 | 4 134 | Apache 2.0 |
| 236 | GPT-4o (2024-05-13) | OpenAI | 1 238 | 1 233–1 244 | 74 322 | Proprietary |
| 237 | Granite 4.1 8B | IBM Granite | 1 237 | 1 223–1 252 | 3 130 | Apache 2.0 |
| 238 | Qwen2.5 Plus | Alibaba Qwen | 1 236 | 1 226–1 246 | 6 369 | Proprietary |
| 239 | Claude 3.5 Sonnet · 20241022 | Anthropic | 1 236 | 1 232–1 240 | 61 432 | Proprietary |
| 240 | DeepSeek V2.5 · v2.5-1210 | DeepSeek | 1 229 | 1 216–1 242 | 4 213 | DeepSeek |
| 241 | Molmo 2 8B | Ai2 | 1 228 | 1 196–1 259 | 572 | Apache 2.0 |
| 242 | Athene V2 Chat | Nexusflow | 1 224 | 1 217–1 231 | 15 433 | NexusFlow |
| 243 | gpt-oss-20b | OpenAI | 1 222 | 1 213–1 231 | 7 601 | Apache 2.0 |
| 244 | GLM-4-Plus · plus | Z.ai | 1 221 | 1 214–1 229 | 16 416 | Proprietary |
| 245 | Gemma 3 4B | 1 221 | 1 207–1 235 | 2 830 | Gemma | |
| 246 | Llama 4 Maverick | Meta | 1 220 | 1 215–1 226 | 28 075 | Llama 4 |
| 247 | Hunyuan Large | Tencent | 1 217 | 1 202–1 232 | 2 472 | Proprietary |
| 248 | Mercury | Inception Labs | 1 216 | 1 195–1 237 | 1 404 | Proprietary |
| 249 | GPT-4o-mini (2024-07-18) | OpenAI | 1 216 | 1 211–1 221 | 43 349 | Proprietary |
| 250 | Gemini 1.5 Flash · 002 | 1 215 | 1 209–1 222 | 22 474 | Proprietary | |
| 251 | Llama 3.1 405B Instruct · bf16 | Meta | 1 213 | 1 208–1 219 | 27 059 | Llama 3.1 Community |
| 252 | Llama 3.1 Nemotron 70B Instruct | NVIDIA | 1 211 | 1 200–1 222 | 4 748 | Llama 3.1 |
| 253 | GPT-4.1 Nano | OpenAI | 1 210 | 1 198–1 222 | 3 981 | Proprietary |
| 254 | GPT-4o (2024-08-06) | OpenAI | 1 209 | 1 202–1 215 | 28 538 | Proprietary |
| 255 | Llama 3.1 405B Instruct · fp8 | Meta | 1 209 | 1 203–1 215 | 38 128 | Llama 3.1 Community |
| 256 | Grok 2 Mini | xAI | 1 208 | 1 202–1 214 | 33 168 | Proprietary |
| 257 | Llama 4 Scout | Meta | 1 208 | 1 201–1 215 | 21 353 | Llama |
| 258 | Claude 3.5 Sonnet · 20240620 | Anthropic | 1 206 | 1 201–1 211 | 55 109 | Proprietary |
| 259 | Qwen Max | Alibaba Qwen | 1 206 | 1 197–1 214 | 10 342 | Qwen |
| 260 | Mistral Small 3.1 24B | Mistral AI | 1 204 | 1 197–1 210 | 23 272 | Apache 2.0 |
| 261 | Gemini 1.5 Pro · advanced-0514 | 1 203 | 1 196–1 210 | 34 077 | Proprietary | |
| 262 | Hunyuan Standard | Tencent | 1 200 | 1 185–1 215 | 2 571 | Proprietary |
| 263 | Llama 3.3 70B Instruct | Meta | 1 199 | 1 194–1 204 | 36 681 | Llama-3.3 |
| 264 | Gemini 1.5 Pro · 1.5-pro-001 | 1 195 | 1 189–1 201 | 52 502 | Proprietary | |
| 265 | DeepSeek V2.5 · v2.5 | DeepSeek | 1 193 | 1 185–1 200 | 15 337 | DeepSeek |
| 266 | GPT-4 Turbo | OpenAI | 1 190 | 1 184–1 196 | 65 114 | Proprietary |
| 267 | Qwen2.5 72B Instruct | Alibaba Qwen | 1 189 | 1 183–1 195 | 24 444 | Qwen |
| 268 | OLMo 3.1 32B Think | Ai2 | 1 188 | 1 177–1 200 | 5 841 | Apache 2.0 |
| 269 | Athene 70B | Nexusflow | 1 188 | 1 179–1 196 | 12 348 | CC-BY-NC-4.0 |
| 270 | Hunyuan Large Vision | Tencent | 1 187 | 1 174–1 200 | 3 931 | Proprietary |
| 271 | Mistral Large 2407 | Mistral AI | 1 185 | 1 179–1 192 | 28 276 | Mistral Research |
| 272 | Mistral Large | Mistral AI | 1 179 | 1 172–1 186 | 18 034 | MRL |
| 273 | Magistral Medium | Mistral AI | 1 178 | 1 169–1 187 | 8 634 | Proprietary |
| 274 | Llama 3.1 70B Instruct | Meta | 1 177 | 1 171–1 182 | 34 840 | Llama 3.1 Community |
| 275 | Claude 3 Opus | Anthropic | 1 175 | 1 170–1 179 | 133 332 | Proprietary |
| 276 | GPT-4 Turbo Preview · 1106-preview | OpenAI | 1 174 | 1 168–1 180 | 69 044 | Proprietary |
| 277 | GPT-4 Turbo Preview · 0125-preview | OpenAI | 1 171 | 1 165–1 178 | 63 519 | Proprietary |
| 278 | Claude 3.5 Haiku | Anthropic | 1 171 | 1 166–1 176 | 48 557 | Proprietary |
| 279 | Nova Pro 1.0 | Amazon | 1 170 | 1 163–1 177 | 15 626 | Proprietary |
| 280 | Llama 3.1 Tulu 3 70B | Ai2 | 1 170 | 1 152–1 187 | 1 728 | Llama 3.1 |
| 281 | Reka Core | Reka AI | 1 156 | 1 145–1 168 | 4 480 | Proprietary |
| 282 | Granite 4.0 H Small | IBM Granite | 1 153 | 1 140–1 166 | 4 110 | Apache 2.0 |
| 283 | Gemini 1.5 Flash · 001 | 1 142 | 1 136–1 149 | 40 479 | Proprietary | |
| 284 | Jamba 1.5 Large | AI21 Labs | 1 139 | 1 128–1 151 | 5 535 | Jamba Open |
| 285 | Mistral Small 3 | Mistral AI | 1 129 | 1 120–1 138 | 9 628 | Apache 2.0 |
| 286 | Gemma 2 27B | 1 129 | 1 123–1 134 | 47 955 | Gemma license | |
| 287 | Qwen2.5 Coder 32B Instruct | Alibaba Qwen | 1 126 | 1 113–1 139 | 3 370 | Apache 2.0 |
| 288 | Command R+ (08-2024) | Cohere | 1 125 | 1 115–1 135 | 6 203 | CC-BY-NC-4.0 |
| 289 | GLM-4 | Z.ai | 1 124 | 1 113–1 134 | 6 229 | Proprietary |
| 290 | Gemma 2 9B IT SimPO | Princeton NLP | 1 122 | 1 112–1 133 | 6 321 | MIT |
| 291 | Llama 3.1 Nemotron 51B Instruct | NVIDIA | 1 122 | 1 106–1 138 | 2 408 | Llama 3.1 |
| 292 | Nemotron-4 340B Instruct | NVIDIA | 1 122 | 1 114–1 130 | 12 726 | NVIDIA Open Model |
| 293 | Nova Lite 1.0 | Amazon | 1 120 | 1 112–1 128 | 12 331 | Proprietary |
| 294 | Gemini 1.5 Flash-8B | 1 118 | 1 111–1 124 | 22 722 | Proprietary | |
| 295 | Llama 3 70B Instruct | Meta | 1 115 | 1 110–1 121 | 103 844 | Llama 3 Community |
| 296 | Aya Expanse 32B | Cohere | 1 115 | 1 107–1 123 | 17 124 | CC-BY-NC-4.0 |
| 297 | Claude 3 Sonnet | Anthropic | 1 108 | 1 102–1 114 | 73 557 | Proprietary |
| 298 | OLMo 2 32B Instruct | Ai2 | 1 107 | 1 090–1 124 | 2 276 | Apache-2.0 |
| 299 | Reka Flash (2024-09) | Reka AI | 1 102 | 1 091–1 114 | 4 562 | Proprietary |
| 300 | Phi 4 | Microsoft | 1 100 | 1 092–1 108 | 15 789 | MIT |
Данные на 13 сентября 2026. Как мы считаем: методика
Свежесть данных
Снимок замеров опубликован 13 сентября 2026; всего снимков борда в архиве: 200.
- замеры оценок людей: загружено 18 сентября 2026, 08:25 UTC
- каталог моделей и цены: загружено 18 сентября 2026, 08:51 UTC
- доли использования: загружено 18 сентября 2026, 08:51 UTC
Другие рейтинги
- Рейтинг агентов
- Рейтинг работы с документами
- Рейтинг редактирования изображений
- Рейтинг видео из изображения
- Популярность моделей по использованию через API-агрегаторы
- Рейтинг поисковых моделей
- Рейтинг моделей для кода
- Рейтинг моделей для творческих текстов
- Рейтинг моделей для математики
- Рейтинг моделей для русского языка
- Рейтинг генерации изображений
- Рейтинг генерации видео
- Рейтинг редактирования видео
- Рейтинг моделей с изображениями на входе
- Рейтинг моделей для веб-разработки
Данные и источники
Обновлено 18 сентября 2026. Каталог, цены и режимы проверяются ежедневно по нескольким API-агрегаторам и провайдерам. Места — по открытым замерам оценок людей, интервалы приведены как в замере.