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: бизнес, управление, финансы», 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 | 1 501 | 1 491–1 511 | 3 920 | Proprietary |
| 2 | Claude Opus 4.6 в режиме high | Anthropic | 1 499 | 1 493–1 505 | 14 403 | Proprietary |
| 3 | Claude Opus 5 в режиме high | Anthropic | 1 499 | 1 491–1 506 | 8 204 | Proprietary |
| 4 | Claude Opus 4.6 | Anthropic | 1 498 | 1 492–1 504 | 15 168 | Proprietary |
| 5 | Muse Spark 1.2 в режиме xhigh | Meta | 1 497 | 1 474–1 521 | 645 | Proprietary |
| 6 | Claude Fable 5.1 в режиме max | Anthropic | 1 494 | 1 476–1 511 | 1 118 | Proprietary |
| 7 | Gemini 3.8 Flash в режиме high | 1 493 | 1 474–1 511 | 1 020 | Proprietary | |
| 8 | Muse Spark 1.3 в режиме max | Meta | 1 492 | 1 473–1 512 | 919 | Proprietary |
| 9 | Claude Opus 4.7 в режиме high | Anthropic | 1 490 | 1 483–1 496 | 12 080 | Proprietary |
| 10 | Claude Fable 5 | Anthropic | 1 481 | 1 473–1 489 | 6 169 | Proprietary |
| 11 | Qwen3.8 Max (0902) | Alibaba Qwen | 1 479 | 1 468–1 490 | 3 217 | Proprietary |
| 12 | Claude Opus 4.7 | Anthropic | 1 478 | 1 471–1 484 | 12 295 | Proprietary |
| 13 | GPT-5.5 в режиме high | OpenAI | 1 478 | 1 471–1 484 | 12 905 | Proprietary |
| 14 | GPT-5.4 в режиме high | OpenAI | 1 476 | 1 470–1 483 | 12 327 | Proprietary |
| 15 | Muse Spark 1.1 | Meta | 1 476 | 1 467–1 484 | 5 528 | Proprietary |
| 16 | ERNIE 5.1 | Baidu | 1 475 | 1 467–1 483 | 7 404 | Proprietary |
| 17 | GPT-5.5 · 5.5 | OpenAI | 1 473 | 1 466–1 479 | 13 208 | Proprietary |
| 18 | Gemini 3.7 Flash в режиме high | 1 472 | 1 453–1 492 | 989 | Proprietary | |
| 19 | Kimi K3 в режиме max | Moonshot AI | 1 468 | 1 458–1 477 | 4 119 | Kimi K3 license |
| 20 | Gemini 3.6 Flash в режиме high | 1 467 | 1 458–1 476 | 5 103 | Proprietary | |
| 21 | MiMo-V2.5-Pro | Xiaomi | 1 467 | 1 460–1 473 | 11 989 | MIT |
| 22 | Gemini 3.5 Flash в режиме high | 1 466 | 1 458–1 473 | 7 618 | Proprietary | |
| 23 | Claude Sonnet 4.6 | Anthropic | 1 465 | 1 459–1 472 | 13 206 | Proprietary |
| 24 | Muse Spark | Meta | 1 463 | 1 451–1 474 | 2 743 | Proprietary |
| 25 | Qwen3.5 Max | Alibaba Qwen | 1 462 | 1 452–1 471 | 4 309 | Proprietary |
| 26 | Gemini 3.1 Pro Preview | 1 459 | 1 454–1 465 | 21 018 | Proprietary | |
| 27 | Gemini 3 Pro | 1 459 | 1 452–1 467 | 7 516 | Proprietary | |
| 28 | Claude Opus 4.8 в режиме high | Anthropic | 1 459 | 1 452–1 466 | 10 607 | Proprietary |
| 29 | GPT-5.6 Sol в режиме xhigh | OpenAI | 1 458 | 1 449–1 466 | 5 232 | Proprietary |
| 30 | GPT-5.4 | OpenAI | 1 457 | 1 451–1 464 | 13 046 | Proprietary |
| 31 | Gemini 3.5 Flash в режиме medium | 1 455 | 1 448–1 463 | 7 279 | Proprietary | |
| 32 | GLM 5.3 в режиме max | Z.ai | 1 455 | 1 442–1 468 | 2 089 | MIT |
| 33 | Claude Opus 4.8 | Anthropic | 1 455 | 1 448–1 462 | 10 600 | Proprietary |
| 34 | GLM 5.3 Flash | Z.ai | 1 455 | 1 441–1 468 | 1 922 | MIT |
| 35 | GLM 5.1 | Z.ai | 1 451 | 1 444–1 457 | 9 676 | MIT |
| 36 | Qwen3.7 Max | Alibaba Qwen | 1 450 | 1 429–1 471 | 835 | Proprietary |
| 37 | Amazon Nova Experimental Chat · 26-02-10 | Amazon | 1 450 | 1 427–1 472 | 605 | Proprietary |
| 38 | Gemini 3 Flash Preview | 1 449 | 1 441–1 457 | 5 580 | Proprietary | |
| 39 | Claude Opus 4.5 | Anthropic | 1 448 | 1 443–1 454 | 13 597 | Proprietary |
| 40 | GPT-5.6 Terra в режиме xhigh | OpenAI | 1 448 | 1 439–1 456 | 5 278 | Proprietary |
| 41 | Kimi K2.6 | Moonshot AI | 1 445 | 1 438–1 453 | 7 473 | Modified MIT |
| 42 | GLM 5.2 в режиме max | Z.ai | 1 444 | 1 437–1 452 | 7 174 | MIT |
| 43 | DeepSeek V4 Pro 0423 | DeepSeek | 1 444 | 1 437–1 451 | 10 721 | MIT |
| 44 | GPT-6 Astra в режиме max | OpenAI | 1 443 | 1 416–1 469 | 502 | Proprietary |
| 45 | Grok 4.5 | xAI | 1 441 | 1 433–1 449 | 5 912 | Proprietary |
| 46 | DeepSeek V4 Pro 0813 в режиме high | DeepSeek | 1 440 | 1 426–1 455 | 1 658 | MIT |
| 47 | Claude Sonnet 4.5 | Anthropic | 1 440 | 1 435–1 446 | 15 295 | Proprietary |
| 48 | Qwen3.8 27B | Alibaba Qwen | 1 440 | 1 426–1 453 | 2 028 | Apache 2.0 |
| 49 | Gemini 2.5 Pro | 1 440 | 1 435–1 444 | 22 313 | Proprietary | |
| 50 | MiMo-V2 Pro | Xiaomi | 1 440 | 1 431–1 449 | 4 832 | Proprietary |
| 51 | GPT-5.6 Luna в режиме xhigh | OpenAI | 1 439 | 1 431–1 448 | 5 508 | Proprietary |
| 52 | Seed 2.0 Pro | ByteDance Seed | 1 439 | 1 433–1 445 | 14 580 | Proprietary |
| 53 | Inkling | Thinking Machines Lab | 1 439 | 1 430–1 447 | 5 051 | Apache 2.0 |
| 54 | Qwen3 Max · preview | Alibaba Qwen | 1 437 | 1 428–1 445 | 5 109 | Proprietary |
| 55 | Claude Sonnet 5 в режиме high | Anthropic | 1 436 | 1 428–1 444 | 7 024 | Proprietary |
| 56 | MiniMax M3 | MiniMax | 1 436 | 1 428–1 443 | 9 608 | MiniMax Community License |
| 57 | Qwen3.5 397B A17B | Alibaba Qwen | 1 434 | 1 428–1 440 | 15 574 | Apache 2.0 |
| 58 | Qwen3.7 Plus | Alibaba Qwen | 1 434 | 1 426–1 441 | 7 922 | Proprietary |
| 59 | GLM 5V Turbo | Z.ai | 1 434 | 1 420–1 447 | 1 947 | Proprietary |
| 60 | Qwen3 VL 235B A22B Instruct | Alibaba Qwen | 1 433 | 1 419–1 447 | 2 209 | Apache 2.0 |
| 61 | Claude Opus 4.5 в режиме high | Anthropic | 1 433 | 1 425–1 441 | 6 806 | Proprietary |
| 62 | Nemotron 3 Ultra | NVIDIA | 1 433 | 1 420–1 446 | 2 140 | OpenMDW-1.1 |
| 63 | Gemma 4 26B A4B | 1 433 | 1 415–1 450 | 1 056 | Apache 2.0 | |
| 64 | Claude Sonnet 4.5 в режиме high | Anthropic | 1 432 | 1 427–1 438 | 15 496 | Proprietary |
| 65 | GPT-5.2 Chat | OpenAI | 1 432 | 1 424–1 440 | 6 666 | Proprietary |
| 66 | Grok 4.20 · beta-0309-reasoning | xAI | 1 432 | 1 425–1 438 | 12 559 | Proprietary |
| 67 | MiMo-V2.5 | Xiaomi | 1 431 | 1 424–1 439 | 9 118 | MIT |
| 68 | Qwen3.6 Plus | Alibaba Qwen | 1 431 | 1 424–1 438 | 9 535 | Proprietary |
| 69 | GPT-5.1 в режиме high | OpenAI | 1 430 | 1 423–1 438 | 7 540 | Proprietary |
| 70 | GLM 5 | Z.ai | 1 430 | 1 422–1 438 | 5 586 | MIT |
| 71 | ERNIE 5.0 · 0110 | Baidu | 1 429 | 1 421–1 436 | 6 735 | Proprietary |
| 72 | LongCat-Flash-Chat · chat | Meituan | 1 428 | 1 416–1 441 | 2 149 | MIT |
| 73 | DeepSeek V4 Pro 0423 в режиме high | DeepSeek | 1 428 | 1 421–1 435 | 10 386 | MIT |
| 74 | GLM 4.7 | Z.ai | 1 428 | 1 416–1 440 | 2 239 | MIT |
| 75 | Kimi K2.5 · thinking | Moonshot AI | 1 428 | 1 422–1 434 | 13 671 | Modified MIT |
| 76 | MiMo-V2 Omni | Xiaomi | 1 428 | 1 418–1 438 | 3 945 | Proprietary |
| 77 | Gemma 4 31B | 1 428 | 1 411–1 445 | 1 146 | Apache 2.0 | |
| 78 | Qwen3.6 Max Preview | Alibaba Qwen | 1 427 | 1 408–1 446 | 1 024 | Proprietary |
| 79 | Mistral Medium 3.5 | Mistral AI | 1 427 | 1 414–1 440 | 2 139 | Modified MIT |
| 80 | Gemini 3.5 Flash Lite | 1 426 | 1 417–1 435 | 4 898 | Proprietary | |
| 81 | Amazon Nova Experimental Chat · 12-10 | Amazon | 1 426 | 1 402–1 449 | 595 | Proprietary |
| 82 | Grok 4.20 · beta1 | xAI | 1 425 | 1 417–1 434 | 5 136 | Proprietary |
| 83 | ChatGPT-4o (latest) | OpenAI | 1 425 | 1 419–1 430 | 13 395 | Proprietary |
| 84 | DeepSeek V4 Flash 0423 | DeepSeek | 1 425 | 1 417–1 432 | 9 891 | MIT |
| 85 | Grok 4.20 Multi-Agent | xAI | 1 424 | 1 417–1 430 | 12 174 | Proprietary |
| 86 | Grok 4.1 · 4.1 | xAI | 1 424 | 1 418–1 430 | 12 677 | Proprietary |
| 87 | Hy3 | Tencent | 1 424 | 1 408–1 439 | 1 587 | Apache 2.0 |
| 88 | Gemini 3 Flash Preview в режиме minimal | 1 423 | 1 418–1 429 | 17 152 | Proprietary | |
| 89 | ERNIE 5.0 · preview-1203 | Baidu | 1 423 | 1 409–1 437 | 1 870 | Proprietary |
| 90 | Qwen3 Next 80B A3B Instruct | Alibaba Qwen | 1 421 | 1 412–1 430 | 4 259 | Apache 2.0 |
| 91 | GLM 4.5 | Z.ai | 1 420 | 1 411–1 429 | 4 225 | MIT |
| 92 | Mistral Large 3 2512 | Mistral AI | 1 419 | 1 414–1 425 | 13 054 | Apache 2.0 |
| 93 | GPT-5.1 | OpenAI | 1 419 | 1 412–1 426 | 8 083 | Proprietary |
| 94 | Amazon Nova Experimental Chat · 11-10 | Amazon | 1 419 | 1 411–1 428 | 4 668 | Proprietary |
| 95 | Mistral Medium 3.1 | Mistral AI | 1 418 | 1 413–1 423 | 17 395 | Proprietary |
| 96 | Qwen3 235B A22B Instruct 2507 | Alibaba Qwen | 1 418 | 1 413–1 423 | 17 814 | Apache 2.0 |
| 97 | Grok 4.1 · 4.1-thinking | xAI | 1 417 | 1 411–1 423 | 12 355 | Proprietary |
| 98 | GPT-5.4 Mini в режиме high | OpenAI | 1 417 | 1 410–1 423 | 12 092 | Proprietary |
| 99 | DeepSeek V3.1 Terminus · terminus-thinking | DeepSeek | 1 417 | 1 395–1 439 | 687 | MIT |
| 100 | GLM 4.6 | Z.ai | 1 417 | 1 409–1 424 | 6 783 | MIT |
| 101 | ERNIE 5.0 · preview-1022 | Baidu | 1 416 | 1 397–1 435 | 922 | Proprietary |
| 102 | Grok 4.6 в режиме high | xAI | 1 416 | 1 405–1 427 | 3 073 | Proprietary |
| 103 | Qwen3.5-122B-A10B | Alibaba Qwen | 1 415 | 1 407–1 423 | 5 466 | Apache 2.0 |
| 104 | GPT-5.2 в режиме high | OpenAI | 1 414 | 1 408–1 421 | 9 375 | Proprietary |
| 105 | Hunyuan Vision 1.5 | Tencent | 1 414 | 1 387–1 441 | 474 | Proprietary |
| 106 | DeepSeek V3.2 Exp · exp | DeepSeek | 1 413 | 1 401–1 426 | 2 201 | MIT |
| 107 | LongCat-Flash-Chat · chat-2602-exp | Meituan | 1 412 | 1 404–1 421 | 5 501 | Proprietary |
| 108 | Qwen3 235B A22B Thinking 2507 | Alibaba Qwen | 1 412 | 1 397–1 427 | 1 525 | Apache 2.0 |
| 109 | DeepSeek V4 Flash 0423 в режиме high | DeepSeek | 1 411 | 1 404–1 418 | 10 056 | MIT |
| 110 | GPT-5.5 · 5.5-instant | OpenAI | 1 410 | 1 401–1 419 | 5 436 | Proprietary |
| 111 | GPT-5.2 | OpenAI | 1 410 | 1 404–1 416 | 15 924 | Proprietary |
| 112 | Grok 3 | xAI | 1 410 | 1 400–1 419 | 3 932 | Proprietary |
| 113 | DeepSeek V3.2 · v3.2 | DeepSeek | 1 410 | 1 403–1 416 | 8 765 | MIT |
| 114 | DeepSeek V3.1 · v3.1-thinking | DeepSeek | 1 409 | 1 396–1 422 | 1 934 | MIT |
| 115 | MiniMax M2.7 | MiniMax | 1 409 | 1 403–1 415 | 13 655 | Modified MIT |
| 116 | DeepSeek V3.2 · v3.2-thinking | DeepSeek | 1 409 | 1 402–1 415 | 7 859 | MIT |
| 117 | Inkling Small | Thinking Machines Lab | 1 407 | 1 397–1 417 | 3 677 | Apache 2.0 |
| 118 | Claude Opus 4.1 · 20250805 | Anthropic | 1 405 | 1 399–1 410 | 13 829 | Proprietary |
| 119 | DeepSeek V3.1 · v3.1 | DeepSeek | 1 405 | 1 393–1 417 | 2 499 | MIT |
| 120 | Claude Opus 4.1 · 20250805-thinking-16k | Anthropic | 1 405 | 1 398–1 411 | 8 972 | Proprietary |
| 121 | Gemini 2.5 Flash · flash | 1 404 | 1 399–1 408 | 22 261 | Proprietary | |
| 122 | Kimi K2.5 · instant | Moonshot AI | 1 402 | 1 387–1 417 | 1 476 | Modified MIT |
| 123 | Amazon Nova Experimental Chat · 10-20 | Amazon | 1 402 | 1 389–1 415 | 2 124 | Proprietary |
| 124 | MiMo-V2 Flash · flash (non-thinking) | Xiaomi | 1 402 | 1 395–1 408 | 8 881 | MIT |
| 125 | R1 0528 | DeepSeek | 1 401 | 1 390–1 413 | 2 845 | MIT |
| 126 | Hy3 preview | Tencent | 1 401 | 1 385–1 417 | 1 382 | tencent-hunyuan-community |
| 127 | Claude Haiku 4.5 | Anthropic | 1 400 | 1 396–1 405 | 25 245 | Proprietary |
| 128 | GPT-5 | OpenAI | 1 399 | 1 391–1 408 | 5 556 | Proprietary |
| 129 | Gemini 2.5 Flash · flash-preview-09-2025 | 1 399 | 1 391–1 406 | 6 050 | Proprietary | |
| 130 | Qwen3.5-27B | Alibaba Qwen | 1 398 | 1 390–1 407 | 5 361 | Apache 2.0 |
| 131 | Muse Glimmer 30B | Meta | 1 398 | 1 376–1 420 | 683 | Apache-2.0 |
| 132 | Kimi K2 Thinking | Moonshot AI | 1 398 | 1 392–1 404 | 11 733 | Modified MIT |
| 133 | Qwen3.5-Flash | Alibaba Qwen | 1 396 | 1 389–1 402 | 11 580 | Proprietary |
| 134 | Qwen3 Max · 2025-09-23 | Alibaba Qwen | 1 395 | 1 381–1 409 | 1 743 | Proprietary |
| 135 | Grok 4 Fast · reasoning | xAI | 1 395 | 1 385–1 405 | 3 528 | Proprietary |
| 136 | DeepSeek V3.2 Exp · exp-thinking | DeepSeek | 1 394 | 1 380–1 408 | 1 768 | MIT |
| 137 | Step 3.5 Flash | StepFun | 1 394 | 1 388–1 400 | 11 219 | Apache 2.0 |
| 138 | Qwen3.5-35B-A3B | Alibaba Qwen | 1 394 | 1 386–1 402 | 5 855 | Apache 2.0 |
| 139 | o3 | OpenAI | 1 392 | 1 386–1 399 | 9 449 | Proprietary |
| 140 | Qwen3 235B A22B · a22b-no-thinking | Alibaba Qwen | 1 392 | 1 384–1 400 | 6 275 | Apache 2.0 |
| 141 | Amazon Nova Experimental Chat · 26-01-10 | Amazon | 1 392 | 1 369–1 415 | 598 | Proprietary |
| 142 | Qwen3 VL 235B A22B Thinking | Alibaba Qwen | 1 391 | 1 377–1 406 | 1 565 | Apache 2.0 |
| 143 | MiniMax M2.1 | MiniMax | 1 390 | 1 379–1 401 | 3 096 | MIT |
| 144 | Gemini 3.1 Flash Lite Preview | 1 390 | 1 384–1 396 | 11 868 | Proprietary | |
| 145 | Qwen3 30B A3B Instruct 2507 | Alibaba Qwen | 1 388 | 1 379–1 397 | 4 102 | Apache 2.0 |
| 146 | Grok 4 | xAI | 1 388 | 1 380–1 395 | 7 032 | Proprietary |
| 147 | DeepSeek V3.1 Terminus · terminus | DeepSeek | 1 387 | 1 365–1 409 | 720 | MIT |
| 148 | MiMo-V2 Flash · flash (thinking) | Xiaomi | 1 386 | 1 373–1 398 | 2 076 | MIT |
| 149 | GPT-4.5 Preview | OpenAI | 1 385 | 1 370–1 401 | 1 394 | Proprietary |
| 150 | Grok 4.1 Fast | xAI | 1 384 | 1 378–1 391 | 10 689 | Proprietary |
| 151 | Hunyuan TurboS · 20250416 | Tencent | 1 383 | 1 368–1 399 | 1 389 | Proprietary |
| 152 | GPT-5.3 Chat | OpenAI | 1 383 | 1 375–1 391 | 6 493 | Proprietary |
| 153 | Grok 4.3 | xAI | 1 381 | 1 375–1 387 | 13 466 | Proprietary |
| 154 | Grok 4 Fast · chat | xAI | 1 380 | 1 363–1 397 | 1 131 | Proprietary |
| 155 | Solar Pro 4 | Upstage | 1 379 | 1 352–1 405 | 493 | Proprietary |
| 156 | GPT-5 в режиме high | OpenAI | 1 378 | 1 369–1 386 | 5 504 | Proprietary |
| 157 | Gemma 3 12B | 1 376 | 1 346–1 406 | 364 | Gemma | |
| 158 | Amazon Nova Experimental Chat · 10-09 | Amazon | 1 376 | 1 352–1 400 | 569 | Proprietary |
| 159 | Hunyuan T1 | Tencent | 1 373 | 1 352–1 394 | 759 | Proprietary |
| 160 | GLM 4.5 Air | Z.ai | 1 371 | 1 363–1 379 | 5 462 | MIT |
| 161 | Nova 2 Lite | Amazon | 1 369 | 1 357–1 382 | 2 268 | Proprietary |
| 162 | GPT-4.1 | OpenAI | 1 369 | 1 362–1 376 | 8 196 | Proprietary |
| 163 | Gemini 2.5 Flash Lite · 09-2025-no-thinking | 1 368 | 1 361–1 374 | 8 935 | Proprietary | |
| 164 | GPT-5.4 Nano в режиме high | OpenAI | 1 365 | 1 358–1 372 | 11 841 | Proprietary |
| 165 | Nemotron 3 Super | NVIDIA | 1 365 | 1 349–1 380 | 1 386 | NVIDIA Open Model |
| 166 | Qwen3 Next 80B A3B Thinking | Alibaba Qwen | 1 361 | 1 349–1 372 | 2 477 | Apache 2.0 |
| 167 | Gemma 3 27B | 1 359 | 1 352–1 367 | 6 476 | Gemma | |
| 168 | gpt-oss-120b | OpenAI | 1 359 | 1 350–1 367 | 5 324 | Apache 2.0 |
| 169 | GPT-5 Mini в режиме high | OpenAI | 1 357 | 1 349–1 366 | 4 795 | Proprietary |
| 170 | GLM 4.6V | Z.ai | 1 357 | 1 332–1 383 | 532 | MIT |
| 171 | Ling-flash-2.0 | inclusionAI | 1 357 | 1 341–1 373 | 1 306 | MIT |
| 172 | MiniMax M2.5 | MiniMax | 1 356 | 1 349–1 363 | 8 021 | Modified MIT |
| 173 | Gemini 2.5 Flash Lite · 06-17-thinking | 1 353 | 1 345–1 362 | 5 410 | Proprietary | |
| 174 | Kimi K2 0905 | Moonshot AI | 1 352 | 1 339–1 365 | 2 046 | Modified MIT |
| 175 | Mercury 2 | Inception Labs | 1 352 | 1 327–1 376 | 479 | Proprietary |
| 176 | Mistral Medium 3 | Mistral AI | 1 351 | 1 343–1 360 | 5 088 | Proprietary |
| 177 | INTELLECT-3 | Prime Intellect | 1 350 | 1 331–1 368 | 986 | MIT |
| 178 | DeepSeek V3 0324 | DeepSeek | 1 349 | 1 342–1 357 | 6 840 | MIT |
| 179 | Qwen2.5 Max | Alibaba Qwen | 1 349 | 1 339–1 359 | 3 657 | Proprietary |
| 180 | Qwen3 235B A22B · a22b | Alibaba Qwen | 1 349 | 1 339–1 358 | 3 915 | Apache 2.0 |
| 181 | Granite 4.2 30B | IBM Granite | 1 349 | 1 324–1 373 | 617 | Apache 2.0 |
| 182 | R1 | DeepSeek | 1 347 | 1 333–1 361 | 1 798 | MIT |
| 183 | Grok 3 Mini | xAI | 1 346 | 1 336–1 357 | 3 328 | Proprietary |
| 184 | Nemotron 3 Nano 30B A3B | NVIDIA | 1 346 | 1 335–1 357 | 2 920 | NVIDIA Open Model |
| 185 | GLM 4.7 Flash | Z.ai | 1 346 | 1 333–1 358 | 2 146 | MIT |
| 186 | Claude Opus 4 · 20250514-thinking-16k | Anthropic | 1 344 | 1 336–1 352 | 5 994 | Proprietary |
| 187 | Qwen3 Coder 480B A35B | Alibaba Qwen | 1 344 | 1 335–1 354 | 4 373 | Apache 2.0 |
| 188 | Grok 3 Mini в режиме high | xAI | 1 344 | 1 332–1 355 | 2 632 | Proprietary |
| 189 | Step 3 | StepFun | 1 343 | 1 327–1 360 | 1 141 | Apache 2.0 |
| 190 | Kimi K2 0711 | Moonshot AI | 1 342 | 1 333–1 352 | 4 538 | Modified MIT |
| 191 | MiniMax M2 | MiniMax | 1 338 | 1 322–1 354 | 1 300 | Apache 2.0 |
| 192 | Claude Opus 4 · 20250514 | Anthropic | 1 337 | 1 329–1 345 | 7 028 | Proprietary |
| 193 | Gemini 2.0 Flash | 1 336 | 1 328–1 345 | 5 259 | Proprietary | |
| 194 | Gemini 2.0 Flash-Lite | 1 335 | 1 323–1 346 | 2 418 | Proprietary | |
| 195 | Trinity Large Thinking | Arcee AI | 1 331 | 1 323–1 340 | 5 846 | Apache 2.0 |
| 196 | Step-1o Turbo | StepFun | 1 329 | 1 313–1 345 | 1 380 | Proprietary |
| 197 | Nemotron 3.5 Lightning | NVIDIA | 1 328 | 1 312–1 344 | 1 561 | OpenMDW-1.1 |
| 198 | o1 · 2024-12-17 | OpenAI | 1 327 | 1 316–1 339 | 2 720 | Proprietary |
| 199 | Trinity Large | Arcee AI | 1 327 | 1 319–1 336 | 5 769 | Apache 2.0 |
| 200 | Ring-flash-2.0 | inclusionAI | 1 326 | 1 310–1 342 | 1 304 | MIT |
| 201 | GLM-4-Plus · plus-0111 | Z.ai | 1 326 | 1 303–1 350 | 592 | Proprietary |
| 202 | Mistral Small 3.2 24B | Mistral AI | 1 325 | 1 314–1 336 | 2 925 | Apache 2.0 |
| 203 | o4 Mini | OpenAI | 1 325 | 1 317–1 332 | 7 015 | Proprietary |
| 204 | Hunyuan TurboS · 20250226 | Tencent | 1 324 | 1 292–1 356 | 263 | Proprietary |
| 205 | Granite 4.2 8B | IBM Granite | 1 323 | 1 298–1 348 | 582 | Apache 2.0 |
| 206 | GPT-4.1 Mini | OpenAI | 1 321 | 1 313–1 329 | 5 999 | Proprietary |
| 207 | QwQ 32B | Alibaba Qwen | 1 320 | 1 310–1 330 | 3 468 | Apache 2.0 |
| 208 | MiniMax M1 | MiniMax | 1 319 | 1 311–1 327 | 5 989 | Apache 2.0 |
| 209 | Claude Sonnet 4 · 20250514 | Anthropic | 1 318 | 1 310–1 326 | 6 513 | Proprietary |
| 210 | Qwen3 32B | Alibaba Qwen | 1 318 | 1 292–1 344 | 437 | Apache 2.0 |
| 211 | OLMo 3.1 32B Instruct | Ai2 | 1 317 | 1 304–1 329 | 2 069 | Apache 2.0 |
| 212 | Command A | Cohere | 1 314 | 1 308–1 321 | 8 747 | CC-BY-NC-4.0 |
| 213 | Claude Sonnet 4 · 20250514-thinking-32k | Anthropic | 1 314 | 1 306–1 322 | 5 758 | Proprietary |
| 214 | o3 Mini High | OpenAI | 1 313 | 1 299–1 326 | 1 815 | Proprietary |
| 215 | Llama 3.3 Nemotron Super 49B v1.5 | NVIDIA | 1 311 | 1 288–1 334 | 593 | Nvidia Open |
| 216 | Gemma 3 4B | 1 311 | 1 281–1 341 | 390 | Gemma | |
| 217 | GPT-5 Nano в режиме high | OpenAI | 1 310 | 1 295–1 326 | 1 523 | Proprietary |
| 218 | Qwen3 30B A3B | Alibaba Qwen | 1 309 | 1 299–1 318 | 3 988 | Apache 2.0 |
| 219 | Hunyuan Turbo | Tencent | 1 308 | 1 275–1 342 | 256 | Proprietary |
| 220 | GLM 4.5V | Z.ai | 1 308 | 1 289–1 327 | 906 | MIT |
| 221 | DeepSeek V3 | DeepSeek | 1 307 | 1 294–1 319 | 2 241 | DeepSeek |
| 222 | o1 · preview | OpenAI | 1 304 | 1 293–1 315 | 3 778 | Proprietary |
| 223 | Qwen-Plus | Alibaba Qwen | 1 303 | 1 280–1 326 | 571 | Proprietary |
| 224 | Llama 3.1 Nemotron Ultra 253B v1 | NVIDIA | 1 301 | 1 267–1 335 | 254 | Nvidia Open Model |
| 225 | Gemma 3n E4B | 1 295 | 1 284–1 305 | 3 268 | Gemma | |
| 226 | Mercury | Inception Labs | 1 293 | 1 261–1 325 | 352 | Proprietary |
| 227 | OLMo 3 32B Think | Ai2 | 1 292 | 1 274–1 310 | 1 068 | Apache 2.0 |
| 228 | o3 Mini | OpenAI | 1 289 | 1 282–1 296 | 7 669 | Proprietary |
| 229 | Qwen2.5 Plus | Alibaba Qwen | 1 289 | 1 271–1 306 | 1 078 | Proprietary |
| 230 | o1-mini | OpenAI | 1 287 | 1 278–1 295 | 6 092 | Proprietary |
| 231 | Llama 3.3 Nemotron Super 49B v1 | NVIDIA | 1 286 | 1 252–1 321 | 241 | Nvidia |
| 232 | Gemini 1.5 Pro · 1.5-pro-002 | 1 286 | 1 277–1 294 | 6 378 | Proprietary | |
| 233 | Granite 4.2 3B | IBM Granite | 1 284 | 1 258–1 310 | 617 | Apache 2.0 |
| 234 | gpt-oss-20b | OpenAI | 1 284 | 1 270–1 298 | 1 835 | Apache 2.0 |
| 235 | Step-2 16k | StepFun | 1 283 | 1 258–1 307 | 506 | Proprietary |
| 236 | Granite 4.1 8B | IBM Granite | 1 276 | 1 252–1 300 | 677 | Apache 2.0 |
| 237 | Athene V2 Chat | Nexusflow | 1 273 | 1 262–1 285 | 2 733 | NexusFlow |
| 238 | Claude 3.7 Sonnet · 20250219 | Anthropic | 1 273 | 1 264–1 281 | 5 745 | Proprietary |
| 239 | GLM-4-Plus · plus | Z.ai | 1 273 | 1 262–1 284 | 3 267 | Proprietary |
| 240 | Yi-Lightning | 01.AI | 1 271 | 1 260–1 282 | 3 420 | Proprietary |
| 241 | Claude 3.7 Sonnet · 20250219-thinking-32k | Anthropic | 1 269 | 1 260–1 277 | 5 581 | Proprietary |
| 242 | GPT-4.1 Nano | OpenAI | 1 266 | 1 244–1 288 | 651 | Proprietary |
| 243 | Grok 2 | xAI | 1 265 | 1 257–1 273 | 7 490 | Proprietary |
| 244 | GPT-4o-mini (2024-07-18) | OpenAI | 1 265 | 1 257–1 272 | 7 765 | Proprietary |
| 245 | Gemini 1.5 Flash · 002 | 1 264 | 1 255–1 274 | 4 259 | Proprietary | |
| 246 | Llama 3.1 Nemotron 70B Instruct | NVIDIA | 1 261 | 1 242–1 280 | 904 | Llama 3.1 |
| 247 | GPT-4o (2024-05-13) | OpenAI | 1 261 | 1 254–1 269 | 13 299 | Proprietary |
| 248 | Llama 4 Maverick | Meta | 1 260 | 1 252–1 268 | 6 125 | Llama 4 |
| 249 | DeepSeek V2.5 · v2.5-1210 | DeepSeek | 1 259 | 1 238–1 280 | 715 | DeepSeek |
| 250 | Mistral Small 3.1 24B | Mistral AI | 1 257 | 1 249–1 266 | 5 503 | Apache 2.0 |
| 251 | Llama 4 Scout | Meta | 1 257 | 1 248–1 266 | 4 917 | Llama |
| 252 | Qwen Max | Alibaba Qwen | 1 256 | 1 243–1 269 | 2 195 | Qwen |
| 253 | Hunyuan Standard | Tencent | 1 255 | 1 226–1 283 | 376 | Proprietary |
| 254 | Claude 3.5 Sonnet · 20241022 | Anthropic | 1 255 | 1 248–1 261 | 11 261 | Proprietary |
| 255 | Grok 2 Mini | xAI | 1 253 | 1 245–1 262 | 6 124 | Proprietary |
| 256 | Qwen2.5 72B Instruct | Alibaba Qwen | 1 253 | 1 243–1 262 | 4 656 | Qwen |
| 257 | OLMo 3.1 32B Think | Ai2 | 1 251 | 1 235–1 267 | 1 467 | Apache 2.0 |
| 258 | Gemini 1.5 Pro · 1.5-pro-001 | 1 248 | 1 239–1 257 | 9 291 | Proprietary | |
| 259 | DeepSeek V2.5 · v2.5 | DeepSeek | 1 248 | 1 236–1 259 | 2 825 | DeepSeek |
| 260 | Llama 3.1 405B Instruct · fp8 | Meta | 1 247 | 1 239–1 256 | 6 667 | Llama 3.1 Community |
| 261 | Gemini 1.5 Pro · advanced-0514 | 1 246 | 1 235–1 256 | 6 130 | Proprietary | |
| 262 | Athene 70B | Nexusflow | 1 245 | 1 232–1 259 | 2 092 | CC-BY-NC-4.0 |
| 263 | Llama 3.1 405B Instruct · bf16 | Meta | 1 245 | 1 236–1 254 | 4 512 | Llama 3.1 Community |
| 264 | Llama 3.3 70B Instruct | Meta | 1 243 | 1 235–1 250 | 6 626 | Llama-3.3 |
| 265 | Hunyuan Large | Tencent | 1 241 | 1 210–1 272 | 357 | Proprietary |
| 266 | Mistral Large | Mistral AI | 1 238 | 1 227–1 248 | 3 021 | MRL |
| 267 | Llama 3.1 70B Instruct | Meta | 1 235 | 1 226–1 243 | 6 381 | Llama 3.1 Community |
| 268 | Claude 3.5 Sonnet · 20240620 | Anthropic | 1 235 | 1 226–1 243 | 9 702 | Proprietary |
| 269 | GPT-4o (2024-08-06) | OpenAI | 1 231 | 1 222–1 241 | 5 095 | Proprietary |
| 270 | Nova Pro 1.0 | Amazon | 1 229 | 1 218–1 240 | 2 591 | Proprietary |
| 271 | Hunyuan Large Vision | Tencent | 1 227 | 1 205–1 249 | 781 | Proprietary |
| 272 | Llama 3.1 Tulu 3 70B | Ai2 | 1 226 | 1 194–1 258 | 288 | Llama 3.1 |
| 273 | Mistral Large 2407 | Mistral AI | 1 226 | 1 216–1 235 | 5 167 | Mistral Research |
| 274 | Gemini 1.5 Flash · 001 | 1 225 | 1 216–1 234 | 7 261 | Proprietary | |
| 275 | Qwen2.5 Coder 32B Instruct | Alibaba Qwen | 1 224 | 1 202–1 245 | 629 | Apache 2.0 |
| 276 | Magistral Medium | Mistral AI | 1 223 | 1 210–1 237 | 2 215 | Proprietary |
| 277 | Claude 3.5 Haiku | Anthropic | 1 222 | 1 215–1 229 | 9 063 | Proprietary |
| 278 | GPT-4 Turbo | OpenAI | 1 222 | 1 214–1 231 | 10 807 | Proprietary |
| 279 | Gemma 2 9B IT SimPO | Princeton NLP | 1 222 | 1 205–1 239 | 1 121 | MIT |
| 280 | Claude 3 Opus | Anthropic | 1 222 | 1 215–1 229 | 21 697 | Proprietary |
| 281 | Nova Lite 1.0 | Amazon | 1 219 | 1 206–1 231 | 2 103 | Proprietary |
| 282 | Reka Core | Reka AI | 1 217 | 1 197–1 236 | 837 | Proprietary |
| 283 | GPT-4 Turbo Preview · 0125-preview | OpenAI | 1 216 | 1 207–1 225 | 9 697 | Proprietary |
| 284 | Llama 3.1 Nemotron 51B Instruct | NVIDIA | 1 215 | 1 188–1 242 | 421 | Llama 3.1 |
| 285 | Command R+ (08-2024) | Cohere | 1 215 | 1 198–1 232 | 1 132 | CC-BY-NC-4.0 |
| 286 | Granite 4.0 H Small | IBM Granite | 1 211 | 1 192–1 231 | 1 037 | Apache 2.0 |
| 287 | GPT-4 Turbo Preview · 1106-preview | OpenAI | 1 211 | 1 202–1 219 | 9 573 | Proprietary |
| 288 | Gemini 1.5 Flash-8B | 1 211 | 1 201–1 220 | 4 291 | Proprietary | |
| 289 | Aya Expanse 32B | Cohere | 1 210 | 1 199–1 221 | 3 115 | CC-BY-NC-4.0 |
| 290 | Gemma 2 27B | 1 207 | 1 200–1 215 | 8 733 | Gemma license | |
| 291 | Mistral Small 3 | Mistral AI | 1 205 | 1 189–1 221 | 1 443 | Apache 2.0 |
| 292 | Phi 4 | Microsoft | 1 198 | 1 186–1 211 | 2 438 | MIT |
| 293 | Llama 3.1 Tulu 3 8B | Ai2 | 1 196 | 1 165–1 228 | 285 | Llama 3.1 |
| 294 | Jamba 1.5 Large | AI21 Labs | 1 196 | 1 178–1 214 | 979 | Jamba Open |
| 295 | Reka Flash (2024-09) | Reka AI | 1 191 | 1 173–1 210 | 904 | Proprietary |
| 296 | Command R+ | Cohere | 1 190 | 1 181–1 200 | 8 504 | CC-BY-NC-4.0 |
| 297 | Gemma 2 9B | 1 190 | 1 181–1 198 | 6 175 | Gemma license | |
| 298 | Nemotron-4 340B Instruct | NVIDIA | 1 185 | 1 171–1 198 | 2 160 | NVIDIA Open Model |
| 299 | Nova Micro 1.0 | Amazon | 1 182 | 1 169–1 195 | 2 061 | Proprietary |
| 300 | OLMo 2 32B Instruct | Ai2 | 1 181 | 1 149–1 213 | 336 | 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.