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: разработка ПО, ИТ-услуги», 300 строк. Оценка и интервал — из источника; интервал не равен диапазону мест.
| Место | Модель | Разработчик | Оценка | Интервал | Голоса | Лицензия |
|---|---|---|---|---|---|---|
| 1 | Claude Opus 4.6 в режиме high | Anthropic | 1 529 | 1 524–1 534 | 27 433 | Proprietary |
| 2 | Claude Opus 4.6 | Anthropic | 1 528 | 1 523–1 532 | 29 751 | Proprietary |
| 3 | Claude Opus 5 в режиме high | Anthropic | 1 527 | 1 521–1 533 | 16 329 | Proprietary |
| 4 | Claude Fable 5.1 в режиме max | Anthropic | 1 526 | 1 512–1 539 | 1 975 | Proprietary |
| 5 | Claude Opus 5 в режиме max | Anthropic | 1 525 | 1 517–1 532 | 7 794 | Proprietary |
| 6 | Muse Spark 1.3 в режиме max | Meta | 1 520 | 1 507–1 534 | 1 943 | Proprietary |
| 7 | Claude Opus 4.7 в режиме high | Anthropic | 1 516 | 1 511–1 521 | 24 222 | Proprietary |
| 8 | Claude Fable 5 | Anthropic | 1 513 | 1 507–1 520 | 11 470 | Proprietary |
| 9 | Claude Opus 4.7 | Anthropic | 1 507 | 1 501–1 512 | 24 473 | Proprietary |
| 10 | Gemini 3.8 Flash в режиме high | 1 505 | 1 492–1 519 | 2 074 | Proprietary | |
| 11 | Muse Spark 1.2 в режиме xhigh | Meta | 1 503 | 1 487–1 519 | 1 321 | Proprietary |
| 12 | GLM 5.3 Flash | Z.ai | 1 503 | 1 493–1 512 | 3 928 | MIT |
| 13 | Gemini 3.7 Flash в режиме high | 1 501 | 1 489–1 514 | 2 327 | Proprietary | |
| 14 | Kimi K3 в режиме max | Moonshot AI | 1 500 | 1 493–1 507 | 8 252 | Kimi K3 license |
| 15 | Qwen3.8 Max (0902) | Alibaba Qwen | 1 499 | 1 491–1 507 | 6 873 | Proprietary |
| 16 | Muse Spark 1.1 | Meta | 1 498 | 1 491–1 504 | 11 002 | Proprietary |
| 17 | GLM 5.3 в режиме max | Z.ai | 1 497 | 1 488–1 507 | 4 225 | MIT |
| 18 | Qwen3.7 Max | Alibaba Qwen | 1 497 | 1 482–1 513 | 1 549 | Proprietary |
| 19 | MiMo-V2.5-Pro | Xiaomi | 1 494 | 1 488–1 499 | 24 174 | MIT |
| 20 | GPT-5.5 в режиме high | OpenAI | 1 493 | 1 488–1 498 | 25 792 | Proprietary |
| 21 | Gemini 3.5 Flash в режиме high | 1 492 | 1 487–1 498 | 15 437 | Proprietary | |
| 22 | ERNIE 5.1 | Baidu | 1 492 | 1 486–1 499 | 14 716 | Proprietary |
| 23 | Gemini 3.6 Flash в режиме high | 1 492 | 1 486–1 499 | 10 864 | Proprietary | |
| 24 | Claude Sonnet 4.6 | Anthropic | 1 492 | 1 487–1 497 | 25 849 | Proprietary |
| 25 | GPT-5.4 в режиме high | OpenAI | 1 491 | 1 486–1 496 | 23 772 | Proprietary |
| 26 | Claude Opus 4.8 в режиме high | Anthropic | 1 488 | 1 482–1 494 | 20 805 | Proprietary |
| 27 | Qwen3.5 Max | Alibaba Qwen | 1 486 | 1 479–1 493 | 8 564 | Proprietary |
| 28 | Gemini 3 Pro | 1 486 | 1 480–1 492 | 13 886 | Proprietary | |
| 29 | Gemini 3.1 Pro Preview | 1 485 | 1 481–1 490 | 42 213 | Proprietary | |
| 30 | Muse Spark | Meta | 1 485 | 1 476–1 493 | 5 318 | Proprietary |
| 31 | GPT-5.5 · 5.5 | OpenAI | 1 484 | 1 479–1 489 | 26 488 | Proprietary |
| 32 | GPT-5.6 Sol в режиме xhigh | OpenAI | 1 484 | 1 477–1 490 | 10 835 | Proprietary |
| 33 | GLM 5.1 | Z.ai | 1 484 | 1 478–1 489 | 19 541 | MIT |
| 34 | Kimi K2.6 | Moonshot AI | 1 483 | 1 477–1 489 | 15 050 | Modified MIT |
| 35 | Claude Opus 4.5 | Anthropic | 1 483 | 1 478–1 487 | 25 768 | Proprietary |
| 36 | Claude Opus 4.8 | Anthropic | 1 482 | 1 477–1 488 | 21 350 | Proprietary |
| 37 | Gemini 3.5 Flash в режиме medium | 1 482 | 1 476–1 488 | 14 840 | Proprietary | |
| 38 | GLM 5.2 в режиме max | Z.ai | 1 481 | 1 475–1 487 | 14 686 | MIT |
| 39 | Claude Opus 4.5 в режиме high | Anthropic | 1 481 | 1 475–1 487 | 12 212 | Proprietary |
| 40 | Amazon Nova Experimental Chat · 26-02-10 | Amazon | 1 481 | 1 465–1 497 | 1 235 | Proprietary |
| 41 | GPT-5.6 Terra в режиме xhigh | OpenAI | 1 478 | 1 472–1 485 | 11 139 | Proprietary |
| 42 | Qwen3.8 27B | Alibaba Qwen | 1 477 | 1 468–1 486 | 4 372 | Apache 2.0 |
| 43 | GPT-5.4 | OpenAI | 1 475 | 1 469–1 480 | 25 245 | Proprietary |
| 44 | Grok 4.5 | xAI | 1 474 | 1 468–1 481 | 12 230 | Proprietary |
| 45 | Seed 2.0 Pro | ByteDance Seed | 1 473 | 1 469–1 478 | 29 095 | Proprietary |
| 46 | Qwen3.7 Plus | Alibaba Qwen | 1 473 | 1 467–1 479 | 16 078 | Proprietary |
| 47 | Claude Sonnet 4.5 | Anthropic | 1 473 | 1 469–1 477 | 29 116 | Proprietary |
| 48 | Claude Sonnet 5 в режиме high | Anthropic | 1 473 | 1 467–1 479 | 14 008 | Proprietary |
| 49 | Gemini 3 Flash Preview | 1 471 | 1 465–1 478 | 10 263 | Proprietary | |
| 50 | Claude Sonnet 4.5 в режиме high | Anthropic | 1 471 | 1 467–1 475 | 29 222 | Proprietary |
| 51 | Kimi K2.5 · instant | Moonshot AI | 1 471 | 1 460–1 482 | 2 751 | Modified MIT |
| 52 | GPT-6 Astra в режиме max | OpenAI | 1 470 | 1 451–1 489 | 1 004 | Proprietary |
| 53 | Kimi K2.5 · thinking | Moonshot AI | 1 469 | 1 465–1 474 | 28 093 | Modified MIT |
| 54 | Nemotron 3 Ultra | NVIDIA | 1 469 | 1 460–1 479 | 4 099 | OpenMDW-1.1 |
| 55 | DeepSeek V4 Pro 0423 | DeepSeek | 1 469 | 1 464–1 474 | 22 255 | MIT |
| 56 | MiMo-V2 Pro | Xiaomi | 1 468 | 1 461–1 475 | 9 822 | Proprietary |
| 57 | MiniMax M3 | MiniMax | 1 467 | 1 461–1 472 | 19 872 | MiniMax Community License |
| 58 | LongCat-Flash-Chat · chat-2602-exp | Meituan | 1 466 | 1 459–1 472 | 10 902 | Proprietary |
| 59 | Inkling | Thinking Machines Lab | 1 465 | 1 458–1 472 | 10 523 | Apache 2.0 |
| 60 | Qwen3.6 Max Preview | Alibaba Qwen | 1 464 | 1 451–1 477 | 2 160 | Proprietary |
| 61 | Hy3 | Tencent | 1 463 | 1 453–1 474 | 3 247 | Apache 2.0 |
| 62 | MiMo-V2.5 | Xiaomi | 1 463 | 1 457–1 469 | 18 133 | MIT |
| 63 | Qwen3.5 397B A17B | Alibaba Qwen | 1 463 | 1 458–1 467 | 31 121 | Apache 2.0 |
| 64 | DeepSeek V4 Pro 0813 в режиме high | DeepSeek | 1 462 | 1 452–1 472 | 3 792 | MIT |
| 65 | ERNIE 5.0 · 0110 | Baidu | 1 462 | 1 456–1 468 | 12 946 | Proprietary |
| 66 | Gemma 4 31B | 1 462 | 1 450–1 474 | 2 135 | Apache 2.0 | |
| 67 | GPT-5.6 Luna в режиме xhigh | OpenAI | 1 462 | 1 455–1 468 | 11 394 | Proprietary |
| 68 | Qwen3.6 Plus | Alibaba Qwen | 1 462 | 1 456–1 467 | 18 914 | Proprietary |
| 69 | Grok 4.20 · beta-0309-reasoning | xAI | 1 461 | 1 456–1 466 | 24 394 | Proprietary |
| 70 | Grok 4.20 Multi-Agent | xAI | 1 461 | 1 456–1 466 | 24 018 | Proprietary |
| 71 | GLM 5V Turbo | Z.ai | 1 460 | 1 450–1 470 | 3 724 | Proprietary |
| 72 | GLM 5 | Z.ai | 1 460 | 1 454–1 466 | 10 510 | MIT |
| 73 | Mistral Medium 3.5 | Mistral AI | 1 459 | 1 450–1 469 | 4 310 | Modified MIT |
| 74 | Qwen3 Max · preview | Alibaba Qwen | 1 459 | 1 452–1 466 | 9 261 | Proprietary |
| 75 | Gemini 2.5 Pro | 1 459 | 1 455–1 463 | 43 018 | Proprietary | |
| 76 | LongCat-Flash-Chat · chat | Meituan | 1 459 | 1 449–1 468 | 3 799 | MIT |
| 77 | Grok 4.6 в режиме high | xAI | 1 458 | 1 450–1 466 | 6 093 | Proprietary |
| 78 | Claude Opus 4.1 · 20250805-thinking-16k | Anthropic | 1 457 | 1 452–1 462 | 16 437 | Proprietary |
| 79 | DeepSeek V4 Pro 0423 в режиме high | DeepSeek | 1 457 | 1 451–1 462 | 21 135 | MIT |
| 80 | MiMo-V2 Omni | Xiaomi | 1 456 | 1 449–1 464 | 7 832 | Proprietary |
| 81 | GLM 4.6 | Z.ai | 1 455 | 1 449–1 461 | 12 303 | MIT |
| 82 | GLM 4.7 | Z.ai | 1 454 | 1 445–1 464 | 3 990 | MIT |
| 83 | GPT-5.2 Chat | OpenAI | 1 454 | 1 448–1 460 | 13 281 | Proprietary |
| 84 | GPT-5.1 в режиме high | OpenAI | 1 454 | 1 448–1 459 | 13 629 | Proprietary |
| 85 | Gemini 3.5 Flash Lite | 1 453 | 1 446–1 460 | 10 628 | Proprietary | |
| 86 | Gemma 4 26B A4B | 1 453 | 1 441–1 465 | 2 106 | Apache 2.0 | |
| 87 | DeepSeek V4 Flash 0423 | DeepSeek | 1 453 | 1 447–1 458 | 20 064 | MIT |
| 88 | Claude Opus 4.1 · 20250805 | Anthropic | 1 452 | 1 447–1 456 | 25 584 | Proprietary |
| 89 | Grok 4.20 · beta1 | xAI | 1 451 | 1 445–1 458 | 10 334 | Proprietary |
| 90 | Grok 4.1 · 4.1-thinking | xAI | 1 451 | 1 446–1 455 | 22 884 | Proprietary |
| 91 | Grok 4.1 · 4.1 | xAI | 1 450 | 1 445–1 455 | 23 959 | Proprietary |
| 92 | Mistral Large 3 2512 | Mistral AI | 1 450 | 1 446–1 454 | 25 886 | Apache 2.0 |
| 93 | DeepSeek V3.2 · v3.2-thinking | DeepSeek | 1 449 | 1 444–1 455 | 13 712 | MIT |
| 94 | ERNIE 5.0 · preview-1203 | Baidu | 1 449 | 1 438–1 459 | 3 241 | Proprietary |
| 95 | Qwen3 Next 80B A3B Instruct | Alibaba Qwen | 1 447 | 1 440–1 454 | 7 934 | Apache 2.0 |
| 96 | Qwen3 235B A22B Instruct 2507 | Alibaba Qwen | 1 447 | 1 443–1 451 | 33 968 | Apache 2.0 |
| 97 | Kimi K2 Thinking | Moonshot AI | 1 446 | 1 441–1 451 | 22 330 | Modified MIT |
| 98 | MiniMax M2.7 | MiniMax | 1 446 | 1 441–1 451 | 28 134 | Modified MIT |
| 99 | Gemini 3 Flash Preview в режиме minimal | 1 446 | 1 441–1 450 | 33 031 | Proprietary | |
| 100 | Amazon Nova Experimental Chat · 12-10 | Amazon | 1 444 | 1 427–1 461 | 1 165 | Proprietary |
| 101 | DeepSeek V3.2 · v3.2 | DeepSeek | 1 443 | 1 438–1 448 | 16 621 | MIT |
| 102 | Qwen3 VL 235B A22B Instruct | Alibaba Qwen | 1 443 | 1 433–1 453 | 3 956 | Apache 2.0 |
| 103 | Inkling Small | Thinking Machines Lab | 1 443 | 1 436–1 450 | 7 696 | Apache 2.0 |
| 104 | DeepSeek V4 Flash 0423 в режиме high | DeepSeek | 1 443 | 1 437–1 448 | 19 996 | MIT |
| 105 | MiMo-V2 Flash · flash (non-thinking) | Xiaomi | 1 442 | 1 437–1 447 | 17 491 | MIT |
| 106 | Amazon Nova Experimental Chat · 11-10 | Amazon | 1 441 | 1 435–1 448 | 8 574 | Proprietary |
| 107 | Amazon Nova Experimental Chat · 26-01-10 | Amazon | 1 440 | 1 423–1 457 | 1 136 | Proprietary |
| 108 | GLM 4.5 | Z.ai | 1 440 | 1 433–1 447 | 8 054 | MIT |
| 109 | Claude Haiku 4.5 | Anthropic | 1 440 | 1 436–1 443 | 48 826 | Proprietary |
| 110 | DeepSeek V3.2 Exp · exp-thinking | DeepSeek | 1 439 | 1 428–1 449 | 3 158 | MIT |
| 111 | Grok 3 | xAI | 1 438 | 1 431–1 445 | 9 097 | Proprietary |
| 112 | Mistral Medium 3.1 | Mistral AI | 1 437 | 1 434–1 441 | 32 982 | Proprietary |
| 113 | Qwen3.5-122B-A10B | Alibaba Qwen | 1 437 | 1 431–1 444 | 11 135 | Apache 2.0 |
| 114 | GPT-5.1 | OpenAI | 1 437 | 1 432–1 443 | 14 847 | Proprietary |
| 115 | Qwen3 Max · 2025-09-23 | Alibaba Qwen | 1 437 | 1 427–1 447 | 3 298 | Proprietary |
| 116 | GPT-5.4 Mini в режиме high | OpenAI | 1 436 | 1 431–1 441 | 23 514 | Proprietary |
| 117 | Step 3.5 Flash | StepFun | 1 435 | 1 430–1 440 | 21 474 | Apache 2.0 |
| 118 | ERNIE 5.0 · preview-1022 | Baidu | 1 435 | 1 421–1 449 | 1 601 | Proprietary |
| 119 | DeepSeek V3.2 Exp · exp | DeepSeek | 1 435 | 1 425–1 444 | 4 034 | MIT |
| 120 | R1 0528 | DeepSeek | 1 435 | 1 426–1 443 | 5 511 | MIT |
| 121 | GPT-5.2 в режиме high | OpenAI | 1 434 | 1 428–1 439 | 17 581 | Proprietary |
| 122 | GPT-5.5 · 5.5-instant | OpenAI | 1 433 | 1 426–1 439 | 10 623 | Proprietary |
| 123 | GPT-5.2 | OpenAI | 1 432 | 1 428–1 437 | 30 446 | Proprietary |
| 124 | Qwen3 235B A22B Thinking 2507 | Alibaba Qwen | 1 431 | 1 420–1 442 | 2 907 | Apache 2.0 |
| 125 | Grok 4 Fast · chat | xAI | 1 431 | 1 418–1 444 | 2 063 | Proprietary |
| 126 | Qwen3.5-27B | Alibaba Qwen | 1 430 | 1 424–1 436 | 10 742 | Apache 2.0 |
| 127 | ChatGPT-4o (latest) | OpenAI | 1 429 | 1 425–1 434 | 26 164 | Proprietary |
| 128 | Hy3 preview | Tencent | 1 428 | 1 416–1 439 | 2 691 | tencent-hunyuan-community |
| 129 | DeepSeek V3.1 · v3.1-thinking | DeepSeek | 1 427 | 1 417–1 438 | 3 471 | MIT |
| 130 | DeepSeek V3.1 · v3.1 | DeepSeek | 1 427 | 1 418–1 436 | 4 636 | MIT |
| 131 | DeepSeek V3.1 Terminus · terminus-thinking | DeepSeek | 1 427 | 1 409–1 444 | 1 131 | MIT |
| 132 | Qwen3 VL 235B A22B Thinking | Alibaba Qwen | 1 426 | 1 415–1 437 | 2 748 | Apache 2.0 |
| 133 | Gemini 2.5 Flash · flash | 1 425 | 1 422–1 429 | 42 771 | Proprietary | |
| 134 | Hunyuan Vision 1.5 | Tencent | 1 422 | 1 402–1 443 | 754 | Proprietary |
| 135 | Amazon Nova Experimental Chat · 10-20 | Amazon | 1 422 | 1 412–1 432 | 3 788 | Proprietary |
| 136 | GPT-5 в режиме high | OpenAI | 1 420 | 1 414–1 426 | 10 682 | Proprietary |
| 137 | MiMo-V2 Flash · flash (thinking) | Xiaomi | 1 419 | 1 410–1 429 | 3 632 | MIT |
| 138 | Solar Pro 4 | Upstage | 1 419 | 1 400–1 438 | 952 | Proprietary |
| 139 | Qwen3.5-Flash | Alibaba Qwen | 1 417 | 1 412–1 422 | 23 550 | Proprietary |
| 140 | Qwen3 30B A3B Instruct 2507 | Alibaba Qwen | 1 417 | 1 410–1 424 | 7 937 | Apache 2.0 |
| 141 | MiniMax M2.1 | MiniMax | 1 417 | 1 409–1 425 | 5 556 | MIT |
| 142 | Grok 4.1 Fast | xAI | 1 417 | 1 412–1 422 | 19 820 | Proprietary |
| 143 | Qwen3.5-35B-A3B | Alibaba Qwen | 1 416 | 1 410–1 422 | 11 448 | Apache 2.0 |
| 144 | Grok 4 Fast · reasoning | xAI | 1 416 | 1 408–1 423 | 6 379 | Proprietary |
| 145 | Grok 4.3 | xAI | 1 415 | 1 410–1 420 | 26 832 | Proprietary |
| 146 | o3 | OpenAI | 1 415 | 1 410–1 420 | 19 035 | Proprietary |
| 147 | Muse Glimmer 30B | Meta | 1 413 | 1 397–1 428 | 1 505 | Apache-2.0 |
| 148 | Gemini 3.1 Flash Lite Preview | 1 413 | 1 407–1 418 | 23 822 | Proprietary | |
| 149 | GPT-4.5 Preview | OpenAI | 1 412 | 1 402–1 423 | 3 358 | Proprietary |
| 150 | Grok 4 | xAI | 1 411 | 1 405–1 417 | 13 389 | Proprietary |
| 151 | Nemotron 3 Super | NVIDIA | 1 411 | 1 400–1 422 | 2 664 | NVIDIA Open Model |
| 152 | DeepSeek V3.1 Terminus · terminus | DeepSeek | 1 410 | 1 394–1 426 | 1 279 | MIT |
| 153 | Hunyuan T1 | Tencent | 1 409 | 1 395–1 424 | 1 513 | Proprietary |
| 154 | Qwen3 235B A22B · a22b-no-thinking | Alibaba Qwen | 1 409 | 1 403–1 415 | 12 387 | Apache 2.0 |
| 155 | Gemini 2.5 Flash · flash-preview-09-2025 | 1 409 | 1 403–1 415 | 11 480 | Proprietary | |
| 156 | Claude Opus 4 · 20250514-thinking-16k | Anthropic | 1 409 | 1 402–1 415 | 11 675 | Proprietary |
| 157 | GPT-5 | OpenAI | 1 408 | 1 402–1 415 | 10 450 | Proprietary |
| 158 | GPT-5.3 Chat | OpenAI | 1 407 | 1 400–1 413 | 12 703 | Proprietary |
| 159 | GPT-5.4 Nano в режиме high | OpenAI | 1 402 | 1 397–1 408 | 23 386 | Proprietary |
| 160 | Ling-flash-2.0 | inclusionAI | 1 401 | 1 389–1 413 | 2 475 | MIT |
| 161 | GLM 4.5 Air | Z.ai | 1 401 | 1 395–1 407 | 10 327 | MIT |
| 162 | Granite 4.2 30B | IBM Granite | 1 400 | 1 383–1 417 | 1 206 | Apache 2.0 |
| 163 | Nova 2 Lite | Amazon | 1 395 | 1 386–1 404 | 4 178 | Proprietary |
| 164 | GPT-5 Mini в режиме high | OpenAI | 1 394 | 1 387–1 401 | 8 816 | Proprietary |
| 165 | Qwen3 Next 80B A3B Thinking | Alibaba Qwen | 1 394 | 1 385–1 402 | 4 681 | Apache 2.0 |
| 166 | Kimi K2 0905 | Moonshot AI | 1 393 | 1 383–1 403 | 3 846 | Modified MIT |
| 167 | Qwen3 Coder 480B A35B | Alibaba Qwen | 1 393 | 1 386–1 400 | 8 426 | Apache 2.0 |
| 168 | GLM 4.7 Flash | Z.ai | 1 391 | 1 382–1 400 | 4 118 | MIT |
| 169 | GPT-4.1 | OpenAI | 1 391 | 1 386–1 397 | 16 312 | Proprietary |
| 170 | Hunyuan TurboS · 20250416 | Tencent | 1 391 | 1 380–1 402 | 3 020 | Proprietary |
| 171 | GLM 4.6V | Z.ai | 1 390 | 1 371–1 409 | 913 | MIT |
| 172 | gpt-oss-120b | OpenAI | 1 388 | 1 382–1 395 | 10 105 | Apache 2.0 |
| 173 | Qwen3 235B A22B · a22b | Alibaba Qwen | 1 386 | 1 379–1 393 | 7 905 | Apache 2.0 |
| 174 | Mistral Medium 3 | Mistral AI | 1 386 | 1 379–1 392 | 10 529 | Proprietary |
| 175 | Mercury 2 | Inception Labs | 1 385 | 1 368–1 403 | 1 054 | Proprietary |
| 176 | Gemini 2.5 Flash Lite · 09-2025-no-thinking | 1 384 | 1 379–1 389 | 16 192 | Proprietary | |
| 177 | Kimi K2 0711 | Moonshot AI | 1 384 | 1 377–1 391 | 8 949 | Modified MIT |
| 178 | Claude Sonnet 4 · 20250514-thinking-32k | Anthropic | 1 383 | 1 377–1 390 | 11 205 | Proprietary |
| 179 | Nemotron 3 Nano 30B A3B | NVIDIA | 1 383 | 1 374–1 391 | 5 117 | NVIDIA Open Model |
| 180 | MiniMax M2.5 | MiniMax | 1 382 | 1 376–1 387 | 15 718 | Modified MIT |
| 181 | Claude Opus 4 · 20250514 | Anthropic | 1 381 | 1 375–1 387 | 14 150 | Proprietary |
| 182 | Grok 3 Mini в режиме high | xAI | 1 379 | 1 371–1 387 | 5 308 | Proprietary |
| 183 | DeepSeek V3 0324 | DeepSeek | 1 379 | 1 373–1 384 | 14 289 | MIT |
| 184 | Amazon Nova Experimental Chat · 10-09 | Amazon | 1 378 | 1 360–1 396 | 967 | Proprietary |
| 185 | Gemini 2.5 Flash Lite · 06-17-thinking | 1 376 | 1 370–1 383 | 10 665 | Proprietary | |
| 186 | R1 | DeepSeek | 1 375 | 1 366–1 384 | 3 954 | MIT |
| 187 | Nemotron 3.5 Lightning | NVIDIA | 1 374 | 1 363–1 385 | 3 273 | OpenMDW-1.1 |
| 188 | MiniMax M2 | MiniMax | 1 374 | 1 362–1 386 | 2 409 | Apache 2.0 |
| 189 | INTELLECT-3 | Prime Intellect | 1 374 | 1 360–1 388 | 1 677 | MIT |
| 190 | Grok 3 Mini | xAI | 1 373 | 1 365–1 380 | 6 868 | Proprietary |
| 191 | Qwen2.5 Max | Alibaba Qwen | 1 373 | 1 366–1 379 | 8 533 | Proprietary |
| 192 | Trinity Large | Arcee AI | 1 370 | 1 364–1 376 | 11 413 | Apache 2.0 |
| 193 | o4 Mini | OpenAI | 1 368 | 1 362–1 373 | 14 604 | Proprietary |
| 194 | o1 · 2024-12-17 | OpenAI | 1 367 | 1 359–1 374 | 6 584 | Proprietary |
| 195 | Qwen3 32B | Alibaba Qwen | 1 366 | 1 347–1 385 | 843 | Apache 2.0 |
| 196 | Step 3 | StepFun | 1 366 | 1 353–1 378 | 2 084 | Apache 2.0 |
| 197 | Ring-flash-2.0 | inclusionAI | 1 365 | 1 354–1 377 | 2 462 | MIT |
| 198 | o3 Mini High | OpenAI | 1 364 | 1 355–1 373 | 4 338 | Proprietary |
| 199 | Trinity Large Thinking | Arcee AI | 1 364 | 1 357–1 370 | 11 734 | Apache 2.0 |
| 200 | o1 · preview | OpenAI | 1 362 | 1 354–1 370 | 8 206 | Proprietary |
| 201 | Gemini 2.0 Flash | 1 362 | 1 356–1 367 | 12 089 | Proprietary | |
| 202 | Mistral Small 3.2 24B | Mistral AI | 1 361 | 1 354–1 369 | 5 891 | Apache 2.0 |
| 203 | Claude Sonnet 4 · 20250514 | Anthropic | 1 361 | 1 355–1 367 | 13 076 | Proprietary |
| 204 | Granite 4.2 8B | IBM Granite | 1 361 | 1 342–1 379 | 1 108 | Apache 2.0 |
| 205 | GPT-4.1 Mini | OpenAI | 1 360 | 1 354–1 366 | 12 375 | Proprietary |
| 206 | MiniMax M1 | MiniMax | 1 359 | 1 353–1 365 | 11 124 | Apache 2.0 |
| 207 | Llama 3.3 Nemotron Super 49B v1.5 | NVIDIA | 1 359 | 1 342–1 376 | 1 179 | Nvidia Open |
| 208 | o1-mini | OpenAI | 1 354 | 1 348–1 360 | 13 214 | Proprietary |
| 209 | Gemma 3 27B | 1 354 | 1 349–1 360 | 13 993 | Gemma | |
| 210 | Step-1o Turbo | StepFun | 1 354 | 1 343–1 365 | 2 780 | Proprietary |
| 211 | GLM 4.5V | Z.ai | 1 354 | 1 340–1 368 | 1 700 | MIT |
| 212 | Granite 4.2 3B | IBM Granite | 1 353 | 1 335–1 372 | 1 146 | Apache 2.0 |
| 213 | Hunyuan TurboS · 20250226 | Tencent | 1 353 | 1 329–1 378 | 470 | Proprietary |
| 214 | OLMo 3.1 32B Instruct | Ai2 | 1 350 | 1 341–1 360 | 3 841 | Apache 2.0 |
| 215 | QwQ 32B | Alibaba Qwen | 1 347 | 1 340–1 354 | 7 330 | Apache 2.0 |
| 216 | GPT-5 Nano в режиме high | OpenAI | 1 345 | 1 334–1 357 | 2 809 | Proprietary |
| 217 | o3 Mini | OpenAI | 1 344 | 1 339–1 349 | 16 564 | Proprietary |
| 218 | Qwen-Plus | Alibaba Qwen | 1 343 | 1 328–1 358 | 1 419 | Proprietary |
| 219 | Qwen3 30B A3B | Alibaba Qwen | 1 340 | 1 333–1 347 | 8 126 | Apache 2.0 |
| 220 | DeepSeek V3 | DeepSeek | 1 337 | 1 328–1 345 | 5 176 | DeepSeek |
| 221 | Command A | Cohere | 1 337 | 1 332–1 342 | 17 839 | CC-BY-NC-4.0 |
| 222 | Gemini 2.0 Flash-Lite | 1 336 | 1 328–1 343 | 5 685 | Proprietary | |
| 223 | OLMo 3 32B Think | Ai2 | 1 333 | 1 319–1 347 | 1 823 | Apache 2.0 |
| 224 | Hunyuan Turbo | Tencent | 1 332 | 1 309–1 356 | 517 | Proprietary |
| 225 | Llama 3.1 Nemotron Ultra 253B v1 | NVIDIA | 1 327 | 1 304–1 350 | 588 | Nvidia Open Model |
| 226 | Mercury | Inception Labs | 1 326 | 1 303–1 348 | 660 | Proprietary |
| 227 | Claude 3.7 Sonnet · 20250219-thinking-32k | Anthropic | 1 324 | 1 318–1 331 | 11 401 | Proprietary |
| 228 | Granite 4.1 8B | IBM Granite | 1 322 | 1 306–1 338 | 1 565 | Apache 2.0 |
| 229 | GLM-4-Plus · plus-0111 | Z.ai | 1 320 | 1 305–1 335 | 1 406 | Proprietary |
| 230 | Qwen2.5 Plus | Alibaba Qwen | 1 317 | 1 305–1 328 | 2 391 | Proprietary |
| 231 | Claude 3.5 Sonnet · 20241022 | Anthropic | 1 317 | 1 312–1 321 | 24 996 | Proprietary |
| 232 | Step-2 16k | StepFun | 1 316 | 1 300–1 331 | 1 200 | Proprietary |
| 233 | Claude 3.7 Sonnet · 20250219 | Anthropic | 1 314 | 1 308–1 320 | 12 707 | Proprietary |
| 234 | Yi-Lightning | 01.AI | 1 314 | 1 306–1 323 | 7 070 | Proprietary |
| 235 | DeepSeek V2.5 · v2.5-1210 | DeepSeek | 1 314 | 1 300–1 328 | 1 688 | DeepSeek |
| 236 | Gemma 3 12B | 1 313 | 1 295–1 332 | 867 | Gemma | |
| 237 | gpt-oss-20b | OpenAI | 1 311 | 1 301–1 322 | 3 589 | Apache 2.0 |
| 238 | Athene V2 Chat | Nexusflow | 1 310 | 1 303–1 318 | 6 182 | NexusFlow |
| 239 | Gemma 3n E4B | 1 308 | 1 300–1 315 | 6 407 | Gemma | |
| 240 | Hunyuan Large | Tencent | 1 307 | 1 287–1 326 | 828 | Proprietary |
| 241 | GPT-4.1 Nano | OpenAI | 1 305 | 1 290–1 320 | 1 454 | Proprietary |
| 242 | Gemini 1.5 Pro · 1.5-pro-002 | 1 304 | 1 298–1 310 | 14 433 | Proprietary | |
| 243 | Molmo 2 8B | Ai2 | 1 304 | 1 269–1 338 | 283 | Apache 2.0 |
| 244 | Mistral Small 3.1 24B | Mistral AI | 1 302 | 1 296–1 308 | 10 844 | Apache 2.0 |
| 245 | Llama 3.3 Nemotron Super 49B v1 | NVIDIA | 1 301 | 1 277–1 325 | 493 | Nvidia |
| 246 | Llama 4 Maverick | Meta | 1 300 | 1 293–1 306 | 12 267 | Llama 4 |
| 247 | OLMo 3.1 32B Think | Ai2 | 1 299 | 1 287–1 311 | 2 597 | Apache 2.0 |
| 248 | GPT-4o (2024-05-13) | OpenAI | 1 298 | 1 292–1 303 | 31 401 | Proprietary |
| 249 | DeepSeek V2.5 · v2.5 | DeepSeek | 1 296 | 1 288–1 303 | 6 692 | DeepSeek |
| 250 | Grok 2 | xAI | 1 295 | 1 289–1 301 | 16 648 | Proprietary |
| 251 | Qwen2.5 72B Instruct | Alibaba Qwen | 1 293 | 1 286–1 299 | 10 448 | Qwen |
| 252 | GLM-4-Plus · plus | Z.ai | 1 292 | 1 284–1 300 | 7 097 | Proprietary |
| 253 | GPT-4o-mini (2024-07-18) | OpenAI | 1 291 | 1 286–1 297 | 17 793 | Proprietary |
| 254 | Qwen Max | Alibaba Qwen | 1 291 | 1 282–1 301 | 4 410 | Qwen |
| 255 | Llama 3.1 405B Instruct · bf16 | Meta | 1 291 | 1 285–1 297 | 10 298 | Llama 3.1 Community |
| 256 | Claude 3.5 Sonnet · 20240620 | Anthropic | 1 289 | 1 283–1 295 | 22 198 | Proprietary |
| 257 | Llama 4 Scout | Meta | 1 285 | 1 279–1 292 | 9 475 | Llama |
| 258 | Hunyuan Standard · 2025-02-10 | Tencent | 1 285 | 1 266–1 304 | 880 | Proprietary |
| 259 | Magistral Medium | Mistral AI | 1 284 | 1 274–1 294 | 3 991 | Proprietary |
| 260 | Llama 3.1 Nemotron 70B Instruct | NVIDIA | 1 284 | 1 271–1 296 | 2 081 | Llama 3.1 |
| 261 | Hunyuan Large Vision | Tencent | 1 281 | 1 267–1 296 | 1 753 | Proprietary |
| 262 | Llama 3.1 405B Instruct · fp8 | Meta | 1 281 | 1 275–1 287 | 15 817 | Llama 3.1 Community |
| 263 | Gemini 1.5 Flash · 002 | 1 279 | 1 272–1 285 | 9 225 | Proprietary | |
| 264 | GPT-4o (2024-08-06) | OpenAI | 1 278 | 1 271–1 285 | 11 864 | Proprietary |
| 265 | Grok 2 Mini | xAI | 1 278 | 1 272–1 284 | 13 865 | Proprietary |
| 266 | Mistral Large 2407 | Mistral AI | 1 277 | 1 270–1 283 | 12 143 | Mistral Research |
| 267 | Mistral Large | Mistral AI | 1 274 | 1 267–1 282 | 6 815 | MRL |
| 268 | Llama 3.3 70B Instruct | Meta | 1 273 | 1 268–1 279 | 14 824 | Llama-3.3 |
| 269 | Gemini 1.5 Pro · 1.5-pro-001 | 1 273 | 1 266–1 279 | 21 409 | Proprietary | |
| 270 | Claude 3.5 Haiku | Anthropic | 1 273 | 1 268–1 278 | 19 584 | Proprietary |
| 271 | Nova Pro 1.0 | Amazon | 1 271 | 1 264–1 279 | 6 163 | Proprietary |
| 272 | Gemma 3 4B | 1 269 | 1 251–1 287 | 986 | Gemma | |
| 273 | GPT-4 Turbo | OpenAI | 1 268 | 1 261–1 274 | 27 573 | Proprietary |
| 274 | Gemini 1.5 Pro · advanced-0514 | 1 267 | 1 259–1 274 | 13 757 | Proprietary | |
| 275 | Athene 70B | Nexusflow | 1 266 | 1 256–1 275 | 5 037 | CC-BY-NC-4.0 |
| 276 | Qwen2.5 Coder 32B Instruct | Alibaba Qwen | 1 261 | 1 247–1 276 | 1 381 | Apache 2.0 |
| 277 | Llama 3.1 70B Instruct | Meta | 1 260 | 1 254–1 266 | 14 811 | Llama 3.1 Community |
| 278 | Claude 3 Opus | Anthropic | 1 259 | 1 254–1 264 | 54 971 | Proprietary |
| 279 | Granite 4.0 H Small | IBM Granite | 1 252 | 1 238–1 265 | 2 099 | Apache 2.0 |
| 280 | GPT-4 Turbo Preview · 1106-preview | OpenAI | 1 251 | 1 244–1 257 | 26 445 | Proprietary |
| 281 | GPT-4 Turbo Preview · 0125-preview | OpenAI | 1 250 | 1 244–1 256 | 25 603 | Proprietary |
| 282 | Mistral Small 3 | Mistral AI | 1 249 | 1 239–1 258 | 3 478 | Apache 2.0 |
| 283 | Llama 3.1 Tulu 3 70B | Ai2 | 1 242 | 1 222–1 262 | 711 | Llama 3.1 |
| 284 | Gemini 1.5 Flash · 001 | 1 242 | 1 235–1 249 | 17 507 | Proprietary | |
| 285 | Nova Lite 1.0 | Amazon | 1 242 | 1 233–1 250 | 4 869 | Proprietary |
| 286 | DeepSeek Coder V2 | DeepSeek | 1 239 | 1 229–1 249 | 4 344 | DeepSeek License |
| 287 | Jamba 1.5 Large | AI21 Labs | 1 237 | 1 225–1 249 | 2 339 | Jamba Open |
| 288 | Reka Core | Reka AI | 1 232 | 1 220–1 245 | 2 006 | Proprietary |
| 289 | Llama 3.1 Nemotron 51B Instruct | NVIDIA | 1 232 | 1 214–1 249 | 1 055 | Llama 3.1 |
| 290 | Gemma 2 27B | 1 228 | 1 223–1 234 | 19 950 | Gemma license | |
| 291 | Phi 4 | Microsoft | 1 228 | 1 220–1 236 | 5 580 | MIT |
| 292 | Gemini 1.5 Flash-8B | 1 228 | 1 221–1 235 | 9 602 | Proprietary | |
| 293 | GLM-4 | Z.ai | 1 226 | 1 215–1 238 | 2 832 | Proprietary |
| 294 | Hunyuan Standard · 256k | Tencent | 1 223 | 1 202–1 243 | 737 | Proprietary |
| 295 | Claude 3 Sonnet | Anthropic | 1 221 | 1 215–1 227 | 31 448 | Proprietary |
| 296 | Nemotron-4 340B Instruct | NVIDIA | 1 216 | 1 207–1 225 | 5 403 | NVIDIA Open Model |
| 297 | Nova Micro 1.0 | Amazon | 1 215 | 1 207–1 224 | 4 685 | Proprietary |
| 298 | OLMo 2 32B Instruct | Ai2 | 1 214 | 1 193–1 236 | 740 | Apache-2.0 |
| 299 | Ministral 8B (2410) | Mistral AI | 1 214 | 1 198–1 229 | 1 252 | MRL |
| 300 | Aya Expanse 32B | Cohere | 1 213 | 1 205–1 221 | 7 302 | CC-BY-NC-4.0 |
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.