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 «немецкий», 299 строк. Оценка и интервал — из источника; интервал не равен диапазону мест.
| Место | Модель | Разработчик | Оценка | Интервал | Голоса | Лицензия |
|---|---|---|---|---|---|---|
| 1 | GLM 5.3 в режиме max | Z.ai | 1 528 | 1 482–1 573 | 155 | MIT |
| 2 | Claude Opus 5 в режиме max | Anthropic | 1 523 | 1 492–1 555 | 368 | Proprietary |
| 3 | Gemini 3 Pro | 1 517 | 1 494–1 539 | 797 | Proprietary | |
| 4 | Claude Opus 4.6 в режиме high | Anthropic | 1 506 | 1 488–1 524 | 1 198 | Proprietary |
| 5 | Claude Opus 5 в режиме high | Anthropic | 1 502 | 1 478–1 526 | 668 | Proprietary |
| 6 | Gemini 3.5 Flash в режиме medium | 1 500 | 1 475–1 525 | 604 | Proprietary | |
| 7 | Gemini 3 Flash Preview | 1 497 | 1 471–1 522 | 564 | Proprietary | |
| 8 | Muse Spark | Meta | 1 495 | 1 456–1 534 | 235 | Proprietary |
| 9 | Gemini 3.1 Pro Preview | 1 491 | 1 476–1 506 | 1 842 | Proprietary | |
| 10 | Claude Opus 4.7 в режиме high | Anthropic | 1 490 | 1 470–1 509 | 965 | Proprietary |
| 11 | Claude Opus 4.6 | Anthropic | 1 488 | 1 470–1 506 | 1 214 | Proprietary |
| 12 | Qwen3.5 Max | Alibaba Qwen | 1 488 | 1 458–1 519 | 371 | Proprietary |
| 13 | Qwen3.8 Max (0902) | Alibaba Qwen | 1 488 | 1 454–1 522 | 266 | Proprietary |
| 14 | Gemini 2.5 Pro | 1 487 | 1 475–1 500 | 2 615 | Proprietary | |
| 15 | Gemini 3.5 Flash в режиме high | 1 485 | 1 461–1 510 | 628 | Proprietary | |
| 16 | Claude Opus 4.7 | Anthropic | 1 483 | 1 464–1 503 | 979 | Proprietary |
| 17 | Claude Fable 5 | Anthropic | 1 483 | 1 456–1 510 | 490 | Proprietary |
| 18 | GPT-5.5 · 5.5 | OpenAI | 1 480 | 1 461–1 500 | 1 046 | Proprietary |
| 19 | Gemini 3.6 Flash в режиме high | 1 474 | 1 445–1 503 | 430 | Proprietary | |
| 20 | GPT-5.6 Sol в режиме xhigh | OpenAI | 1 473 | 1 445–1 502 | 426 | Proprietary |
| 21 | Qwen3.7 Plus | Alibaba Qwen | 1 473 | 1 451–1 496 | 719 | Proprietary |
| 22 | Claude Opus 4.8 в режиме high | Anthropic | 1 473 | 1 453–1 493 | 917 | Proprietary |
| 23 | Gemini 3 Flash Preview в режиме minimal | 1 471 | 1 454–1 488 | 1 430 | Proprietary | |
| 24 | GPT-5.4 в режиме high | OpenAI | 1 470 | 1 451–1 489 | 1 050 | Proprietary |
| 25 | GLM 5.1 | Z.ai | 1 468 | 1 446–1 489 | 788 | MIT |
| 26 | GLM 5.2 в режиме max | Z.ai | 1 468 | 1 445–1 491 | 686 | MIT |
| 27 | ERNIE 5.1 | Baidu | 1 467 | 1 445–1 489 | 723 | Proprietary |
| 28 | Qwen3 Max · preview | Alibaba Qwen | 1 465 | 1 440–1 490 | 595 | Proprietary |
| 29 | Nemotron 3 Ultra | NVIDIA | 1 463 | 1 423–1 503 | 201 | OpenMDW-1.1 |
| 30 | Grok 4.20 · beta1 | xAI | 1 463 | 1 435–1 491 | 435 | Proprietary |
| 31 | Kimi K3 в режиме max | Moonshot AI | 1 463 | 1 433–1 492 | 413 | Kimi K3 license |
| 32 | GPT-5.5 в режиме high | OpenAI | 1 462 | 1 442–1 481 | 1 014 | Proprietary |
| 33 | ERNIE 5.0 · preview-1203 | Baidu | 1 460 | 1 419–1 501 | 220 | Proprietary |
| 34 | Muse Spark 1.1 | Meta | 1 459 | 1 433–1 486 | 510 | Proprietary |
| 35 | GPT-5.6 Luna в режиме xhigh | OpenAI | 1 459 | 1 432–1 486 | 511 | Proprietary |
| 36 | Grok 4.20 Multi-Agent | xAI | 1 458 | 1 438–1 478 | 1 023 | Proprietary |
| 37 | GPT-5.6 Terra в режиме xhigh | OpenAI | 1 457 | 1 430–1 484 | 461 | Proprietary |
| 38 | GPT-4.5 Preview | OpenAI | 1 456 | 1 424–1 489 | 319 | Proprietary |
| 39 | DeepSeek V4 Pro 0423 | DeepSeek | 1 456 | 1 437–1 476 | 955 | MIT |
| 40 | MiMo-V2.5-Pro | Xiaomi | 1 455 | 1 436–1 475 | 1 048 | MIT |
| 41 | Gemini 3.5 Flash Lite | 1 454 | 1 427–1 481 | 485 | Proprietary | |
| 42 | Qwen3.6 Plus | Alibaba Qwen | 1 452 | 1 430–1 474 | 758 | Proprietary |
| 43 | Claude Opus 4.8 | Anthropic | 1 451 | 1 431–1 472 | 919 | Proprietary |
| 44 | Claude Sonnet 5 в режиме high | Anthropic | 1 450 | 1 425–1 475 | 586 | Proprietary |
| 45 | Kimi K2.6 | Moonshot AI | 1 449 | 1 427–1 471 | 724 | Modified MIT |
| 46 | Grok 4.5 | xAI | 1 449 | 1 423–1 475 | 512 | Proprietary |
| 47 | Qwen3.8 27B | Alibaba Qwen | 1 447 | 1 402–1 492 | 160 | Apache 2.0 |
| 48 | GPT-5.2 в режиме high | OpenAI | 1 447 | 1 426–1 468 | 817 | Proprietary |
| 49 | Claude Opus 4.5 | Anthropic | 1 446 | 1 428–1 465 | 1 104 | Proprietary |
| 50 | Grok 4.1 · 4.1 | xAI | 1 446 | 1 428–1 463 | 1 233 | Proprietary |
| 51 | GLM 4.6 | Z.ai | 1 445 | 1 423–1 468 | 677 | MIT |
| 52 | DeepSeek V4 Pro 0423 в режиме high | DeepSeek | 1 445 | 1 425–1 466 | 879 | MIT |
| 53 | Qwen3.5 397B A17B | Alibaba Qwen | 1 445 | 1 427–1 463 | 1 220 | Apache 2.0 |
| 54 | GLM 5.3 Flash | Z.ai | 1 444 | 1 399–1 489 | 160 | MIT |
| 55 | GPT-5.2 Chat | OpenAI | 1 444 | 1 418–1 469 | 553 | Proprietary |
| 56 | GLM 5 | Z.ai | 1 444 | 1 417–1 470 | 497 | MIT |
| 57 | DeepSeek V3.2 Exp · exp | DeepSeek | 1 443 | 1 406–1 479 | 226 | MIT |
| 58 | GPT-5.4 | OpenAI | 1 441 | 1 421–1 461 | 955 | Proprietary |
| 59 | Kimi K2.5 · thinking | Moonshot AI | 1 441 | 1 423–1 459 | 1 247 | Modified MIT |
| 60 | Claude Opus 4.5 в режиме high | Anthropic | 1 441 | 1 417–1 464 | 656 | Proprietary |
| 61 | Grok 4.20 · beta-0309-reasoning | xAI | 1 441 | 1 421–1 460 | 1 019 | Proprietary |
| 62 | GPT-5.1 в режиме high | OpenAI | 1 440 | 1 418–1 462 | 703 | Proprietary |
| 63 | Grok 4.1 · 4.1-thinking | xAI | 1 440 | 1 421–1 458 | 1 128 | Proprietary |
| 64 | DeepSeek V3.2 Exp · exp-thinking | DeepSeek | 1 439 | 1 396–1 481 | 183 | MIT |
| 65 | Seed 2.0 Pro | ByteDance Seed | 1 438 | 1 420–1 456 | 1 282 | Proprietary |
| 66 | Mistral Large 3 2512 | Mistral AI | 1 437 | 1 418–1 455 | 1 153 | Apache 2.0 |
| 67 | ERNIE 5.0 · 0110 | Baidu | 1 436 | 1 412–1 459 | 651 | Proprietary |
| 68 | Qwen3 Max · 2025-09-23 | Alibaba Qwen | 1 435 | 1 395–1 475 | 204 | Proprietary |
| 69 | Inkling | Thinking Machines Lab | 1 435 | 1 406–1 464 | 407 | Apache 2.0 |
| 70 | Mistral Medium 3.1 | Mistral AI | 1 432 | 1 418–1 447 | 1 724 | Proprietary |
| 71 | Qwen3 235B A22B Instruct 2507 | Alibaba Qwen | 1 432 | 1 417–1 446 | 1 788 | Apache 2.0 |
| 72 | Inkling Small | Thinking Machines Lab | 1 431 | 1 397–1 466 | 297 | Apache 2.0 |
| 73 | Grok 4.6 в режиме high | xAI | 1 431 | 1 395–1 466 | 287 | Proprietary |
| 74 | Grok 3 | xAI | 1 431 | 1 409–1 452 | 783 | Proprietary |
| 75 | GLM 4.7 | Z.ai | 1 428 | 1 391–1 466 | 219 | MIT |
| 76 | GPT-5.2 | OpenAI | 1 428 | 1 410–1 446 | 1 225 | Proprietary |
| 77 | Grok 4 | xAI | 1 428 | 1 408–1 448 | 898 | Proprietary |
| 78 | Gemini 3.1 Flash Lite Preview | 1 427 | 1 408–1 446 | 1 013 | Proprietary | |
| 79 | Claude Sonnet 4.5 в режиме high | Anthropic | 1 426 | 1 409–1 442 | 1 331 | Proprietary |
| 80 | MiniMax M3 | MiniMax | 1 426 | 1 405–1 447 | 881 | MiniMax Community License |
| 81 | Qwen3.5-122B-A10B | Alibaba Qwen | 1 423 | 1 395–1 451 | 418 | Apache 2.0 |
| 82 | ChatGPT-4o (latest) | OpenAI | 1 423 | 1 408–1 437 | 1 763 | Proprietary |
| 83 | Claude Sonnet 4.6 | Anthropic | 1 423 | 1 404–1 441 | 1 161 | Proprietary |
| 84 | Qwen3 VL 235B A22B Instruct | Alibaba Qwen | 1 422 | 1 387–1 458 | 278 | Apache 2.0 |
| 85 | Claude Sonnet 4.5 | Anthropic | 1 421 | 1 405–1 438 | 1 336 | Proprietary |
| 86 | o3 | OpenAI | 1 420 | 1 404–1 436 | 1 507 | Proprietary |
| 87 | Gemini 2.5 Flash · flash-preview-09-2025 | 1 419 | 1 395–1 442 | 648 | Proprietary | |
| 88 | MiMo-V2 Pro | Xiaomi | 1 418 | 1 390–1 446 | 434 | Proprietary |
| 89 | DeepSeek V4 Flash 0423 | DeepSeek | 1 418 | 1 396–1 439 | 822 | MIT |
| 90 | GPT-5.1 | OpenAI | 1 418 | 1 396–1 440 | 735 | Proprietary |
| 91 | Qwen3 Next 80B A3B Instruct | Alibaba Qwen | 1 417 | 1 391–1 444 | 487 | Apache 2.0 |
| 92 | Gemini 2.5 Flash · flash | 1 417 | 1 405–1 429 | 2 665 | Proprietary | |
| 93 | Qwen3 VL 235B A22B Thinking | Alibaba Qwen | 1 417 | 1 373–1 460 | 179 | Apache 2.0 |
| 94 | Kimi K2.5 · instant | Moonshot AI | 1 414 | 1 368–1 461 | 141 | Modified MIT |
| 95 | GPT-5 в режиме high | OpenAI | 1 414 | 1 391–1 436 | 743 | Proprietary |
| 96 | DeepSeek V4 Flash 0423 в режиме high | DeepSeek | 1 413 | 1 392–1 434 | 866 | MIT |
| 97 | Claude Opus 4.1 · 20250805-thinking-16k | Anthropic | 1 413 | 1 394–1 432 | 1 041 | Proprietary |
| 98 | DeepSeek V3.2 · v3.2-thinking | DeepSeek | 1 411 | 1 390–1 432 | 736 | MIT |
| 99 | MiMo-V2.5 | Xiaomi | 1 411 | 1 389–1 433 | 749 | MIT |
| 100 | DeepSeek V3.1 · v3.1 | DeepSeek | 1 411 | 1 381–1 440 | 397 | MIT |
| 101 | DeepSeek V3.1 · v3.1-thinking | DeepSeek | 1 409 | 1 377–1 441 | 324 | MIT |
| 102 | LongCat-Flash-Chat · chat | Meituan | 1 408 | 1 374–1 443 | 297 | MIT |
| 103 | GPT-5.4 Mini в режиме high | OpenAI | 1 408 | 1 388–1 429 | 890 | Proprietary |
| 104 | GPT-5.5 · 5.5-instant | OpenAI | 1 408 | 1 379–1 437 | 424 | Proprietary |
| 105 | DeepSeek V3.2 · v3.2 | DeepSeek | 1 408 | 1 387–1 428 | 892 | MIT |
| 106 | Qwen3 235B A22B Thinking 2507 | Alibaba Qwen | 1 406 | 1 367–1 446 | 210 | Apache 2.0 |
| 107 | GLM 4.5 | Z.ai | 1 406 | 1 381–1 432 | 537 | MIT |
| 108 | Claude Opus 4.1 · 20250805 | Anthropic | 1 406 | 1 391–1 422 | 1 551 | Proprietary |
| 109 | GPT-5 | OpenAI | 1 405 | 1 382–1 427 | 699 | Proprietary |
| 110 | R1 0528 | DeepSeek | 1 404 | 1 380–1 429 | 592 | MIT |
| 111 | Step 3.5 Flash | StepFun | 1 403 | 1 384–1 422 | 1 045 | Apache 2.0 |
| 112 | Grok 4.1 Fast | xAI | 1 402 | 1 382–1 422 | 920 | Proprietary |
| 113 | MiMo-V2 Omni | Xiaomi | 1 397 | 1 365–1 428 | 347 | Proprietary |
| 114 | Amazon Nova Experimental Chat · 11-10 | Amazon | 1 396 | 1 369–1 423 | 471 | Proprietary |
| 115 | Qwen3.5-27B | Alibaba Qwen | 1 396 | 1 368–1 423 | 440 | Apache 2.0 |
| 116 | Grok 4.3 | xAI | 1 395 | 1 375–1 414 | 996 | Proprietary |
| 117 | MiniMax M2.7 | MiniMax | 1 393 | 1 374–1 411 | 1 201 | Modified MIT |
| 118 | Kimi K2 Thinking | Moonshot AI | 1 393 | 1 374–1 411 | 1 126 | Modified MIT |
| 119 | Gemini 2.5 Flash Lite · 09-2025-no-thinking | 1 390 | 1 370–1 411 | 832 | Proprietary | |
| 120 | Amazon Nova Experimental Chat · 10-20 | Amazon | 1 390 | 1 356–1 425 | 292 | Proprietary |
| 121 | Qwen3.5-Flash | Alibaba Qwen | 1 390 | 1 371–1 409 | 1 028 | Proprietary |
| 122 | Claude Opus 4 · 20250514-thinking-16k | Anthropic | 1 390 | 1 370–1 409 | 951 | Proprietary |
| 123 | MiMo-V2 Flash · flash (thinking) | Xiaomi | 1 388 | 1 351–1 425 | 254 | MIT |
| 124 | Qwen3 235B A22B · a22b | Alibaba Qwen | 1 388 | 1 366–1 410 | 758 | Apache 2.0 |
| 125 | Kimi K2 0905 | Moonshot AI | 1 386 | 1 351–1 421 | 276 | Modified MIT |
| 126 | R1 | DeepSeek | 1 384 | 1 356–1 412 | 436 | MIT |
| 127 | Grok 4 Fast | xAI | 1 383 | 1 352–1 414 | 358 | Proprietary |
| 128 | Qwen3 235B A22B · a22b-no-thinking | Alibaba Qwen | 1 383 | 1 364–1 401 | 1 074 | Apache 2.0 |
| 129 | MiMo-V2 Flash · flash (non-thinking) | Xiaomi | 1 381 | 1 360–1 403 | 776 | MIT |
| 130 | GPT-4.1 | OpenAI | 1 381 | 1 364–1 397 | 1 278 | Proprietary |
| 131 | Qwen3 30B A3B Instruct 2507 | Alibaba Qwen | 1 380 | 1 354–1 406 | 522 | Apache 2.0 |
| 132 | GPT-5.3 Chat | OpenAI | 1 380 | 1 354–1 406 | 513 | Proprietary |
| 133 | MiniMax M2.1 | MiniMax | 1 379 | 1 343–1 414 | 293 | MIT |
| 134 | Claude Haiku 4.5 | Anthropic | 1 378 | 1 364–1 392 | 2 213 | Proprietary |
| 135 | Mistral Medium 3 | Mistral AI | 1 377 | 1 358–1 396 | 991 | Proprietary |
| 136 | GLM 4.5 Air | Z.ai | 1 374 | 1 352–1 396 | 735 | MIT |
| 137 | Kimi K2 0711 | Moonshot AI | 1 374 | 1 351–1 397 | 648 | Modified MIT |
| 138 | DeepSeek V3 0324 | DeepSeek | 1 374 | 1 356–1 391 | 1 220 | MIT |
| 139 | Gemini 2.5 Flash Lite · 06-17-thinking | 1 373 | 1 353–1 394 | 841 | Proprietary | |
| 140 | Claude Opus 4 · 20250514 | Anthropic | 1 372 | 1 355–1 389 | 1 275 | Proprietary |
| 141 | Step 3 | StepFun | 1 371 | 1 325–1 417 | 152 | Apache 2.0 |
| 142 | Gemma 3 12B | 1 370 | 1 328–1 413 | 166 | Gemma | |
| 143 | LongCat-Flash-Chat · chat-2602-exp | Meituan | 1 370 | 1 342–1 397 | 437 | Proprietary |
| 144 | GPT-5.4 Nano в режиме high | OpenAI | 1 368 | 1 347–1 389 | 909 | Proprietary |
| 145 | GPT-5 Mini в режиме high | OpenAI | 1 365 | 1 341–1 389 | 585 | Proprietary |
| 146 | Qwen3.5-35B-A3B | Alibaba Qwen | 1 363 | 1 337–1 389 | 478 | Apache 2.0 |
| 147 | MiniMax M2 | MiniMax | 1 362 | 1 309–1 415 | 128 | Apache 2.0 |
| 148 | MiniMax M2.5 | MiniMax | 1 362 | 1 337–1 386 | 627 | Modified MIT |
| 149 | Gemma 3 27B | 1 361 | 1 343–1 378 | 1 204 | Gemma | |
| 150 | Qwen3 Next 80B A3B Thinking | Alibaba Qwen | 1 357 | 1 324–1 390 | 295 | Apache 2.0 |
| 151 | Hunyuan TurboS | Tencent | 1 356 | 1 323–1 388 | 328 | Proprietary |
| 152 | Trinity Large Thinking | Arcee AI | 1 355 | 1 326–1 383 | 406 | Apache 2.0 |
| 153 | gpt-oss-120b | OpenAI | 1 353 | 1 331–1 376 | 666 | Apache 2.0 |
| 154 | Gemini 2.0 Flash | 1 353 | 1 336–1 371 | 1 065 | Proprietary | |
| 155 | Nemotron 3 Nano 30B A3B | NVIDIA | 1 350 | 1 316–1 385 | 291 | NVIDIA Open Model |
| 156 | Qwen2.5 Max | Alibaba Qwen | 1 349 | 1 328–1 370 | 781 | Proprietary |
| 157 | Grok 3 Mini в режиме high | xAI | 1 349 | 1 320–1 378 | 415 | Proprietary |
| 158 | GPT-4.1 Mini | OpenAI | 1 349 | 1 331–1 368 | 1 050 | Proprietary |
| 159 | MiniMax M1 | MiniMax | 1 349 | 1 329–1 370 | 876 | Apache 2.0 |
| 160 | Grok 3 Mini | xAI | 1 347 | 1 322–1 372 | 537 | Proprietary |
| 161 | Nova 2 Lite | Amazon | 1 346 | 1 311–1 382 | 280 | Proprietary |
| 162 | GLM 4.7 Flash | Z.ai | 1 341 | 1 301–1 382 | 217 | MIT |
| 163 | Command A | Cohere | 1 341 | 1 325–1 357 | 1 363 | CC-BY-NC-4.0 |
| 164 | Mistral Small 3.2 24B | Mistral AI | 1 341 | 1 312–1 370 | 426 | Apache 2.0 |
| 165 | Qwen3 32B | Alibaba Qwen | 1 341 | 1 300–1 382 | 172 | Apache 2.0 |
| 166 | Nemotron 3 Super | NVIDIA | 1 341 | 1 300–1 382 | 197 | NVIDIA Open Model |
| 167 | GLM-4-Plus · plus-0111 | Z.ai | 1 341 | 1 301–1 380 | 207 | Proprietary |
| 168 | o1 · 2024-12-17 | OpenAI | 1 337 | 1 313–1 361 | 556 | Proprietary |
| 169 | o4 Mini | OpenAI | 1 336 | 1 318–1 353 | 1 212 | Proprietary |
| 170 | Claude Sonnet 4 · 20250514 | Anthropic | 1 330 | 1 312–1 349 | 1 077 | Proprietary |
| 171 | GPT-5 Nano в режиме high | OpenAI | 1 328 | 1 287–1 370 | 204 | Proprietary |
| 172 | DeepSeek V3 | DeepSeek | 1 323 | 1 298–1 348 | 493 | DeepSeek |
| 173 | Qwen3 Coder 480B A35B | Alibaba Qwen | 1 323 | 1 298–1 349 | 588 | Apache 2.0 |
| 174 | Claude Sonnet 4 · 20250514-thinking-32k | Anthropic | 1 318 | 1 297–1 339 | 860 | Proprietary |
| 175 | Trinity Large | Arcee AI | 1 317 | 1 288–1 346 | 458 | Apache 2.0 |
| 176 | Step-1o Turbo | StepFun | 1 314 | 1 283–1 346 | 345 | Proprietary |
| 177 | QwQ 32B | Alibaba Qwen | 1 313 | 1 290–1 335 | 668 | Apache 2.0 |
| 178 | Gemma 3n E4B | 1 311 | 1 290–1 333 | 719 | Gemma | |
| 179 | o1 · preview | OpenAI | 1 309 | 1 287–1 331 | 774 | Proprietary |
| 180 | Qwen3 30B A3B | Alibaba Qwen | 1 307 | 1 286–1 329 | 764 | Apache 2.0 |
| 181 | Gemini 2.0 Flash-Lite | 1 305 | 1 283–1 328 | 627 | Proprietary | |
| 182 | o3 Mini High | OpenAI | 1 303 | 1 273–1 332 | 377 | Proprietary |
| 183 | Claude 3.7 Sonnet · 20250219-thinking-32k | Anthropic | 1 301 | 1 281–1 320 | 1 022 | Proprietary |
| 184 | o3 Mini | OpenAI | 1 299 | 1 282–1 315 | 1 341 | Proprietary |
| 185 | Claude 3.5 Sonnet · 20241022 | Anthropic | 1 297 | 1 284–1 311 | 2 021 | Proprietary |
| 186 | OLMo 3 32B Think | Ai2 | 1 292 | 1 239–1 346 | 135 | Apache 2.0 |
| 187 | Llama 4 Maverick | Meta | 1 291 | 1 272–1 310 | 1 026 | Llama 4 |
| 188 | GPT-4.1 Nano | OpenAI | 1 288 | 1 251–1 324 | 222 | Proprietary |
| 189 | Grok 2 | xAI | 1 287 | 1 271–1 302 | 1 609 | Proprietary |
| 190 | Gemini 1.5 Pro · 1.5-pro-002 | 1 286 | 1 269–1 303 | 1 281 | Proprietary | |
| 191 | OLMo 3.1 32B Instruct | Ai2 | 1 282 | 1 245–1 320 | 242 | Apache 2.0 |
| 192 | GPT-4o (2024-05-13) | OpenAI | 1 282 | 1 269–1 296 | 3 573 | Proprietary |
| 193 | Claude 3.7 Sonnet · 20250219 | Anthropic | 1 281 | 1 262–1 300 | 1 082 | Proprietary |
| 194 | Gemma 3 4B | 1 280 | 1 238–1 323 | 162 | Gemma | |
| 195 | GLM-4-Plus · plus | Z.ai | 1 278 | 1 255–1 301 | 659 | Proprietary |
| 196 | o1-mini | OpenAI | 1 278 | 1 260–1 296 | 1 169 | Proprietary |
| 197 | Gemini 1.5 Pro · advanced-0514 | 1 275 | 1 257–1 292 | 1 779 | Proprietary | |
| 198 | Llama 4 Scout | Meta | 1 273 | 1 251–1 295 | 794 | Llama |
| 199 | Grok 2 Mini | xAI | 1 273 | 1 255–1 290 | 1 298 | Proprietary |
| 200 | GPT-4o-mini (2024-07-18) | OpenAI | 1 272 | 1 257–1 287 | 1 641 | Proprietary |
| 201 | Claude 3.5 Sonnet · 20240620 | Anthropic | 1 271 | 1 256–1 285 | 2 499 | Proprietary |
| 202 | Yi-Lightning | 01.AI | 1 267 | 1 245–1 288 | 700 | Proprietary |
| 203 | Mistral Small 3.1 24B | Mistral AI | 1 266 | 1 245–1 287 | 873 | Apache 2.0 |
| 204 | OLMo 3.1 32B Think | Ai2 | 1 262 | 1 213–1 310 | 163 | Apache 2.0 |
| 205 | Gemini 1.5 Flash · 002 | 1 261 | 1 240–1 283 | 739 | Proprietary | |
| 206 | Athene V2 Chat | Nexusflow | 1 260 | 1 238–1 283 | 626 | NexusFlow |
| 207 | GPT-4 Turbo | OpenAI | 1 259 | 1 245–1 273 | 3 359 | Proprietary |
| 208 | Claude 3 Opus | Anthropic | 1 258 | 1 246–1 271 | 5 672 | Proprietary |
| 209 | DeepSeek V2.5 · v2.5-1210 | DeepSeek | 1 258 | 1 212–1 304 | 140 | DeepSeek |
| 210 | GPT-4o (2024-08-06) | OpenAI | 1 257 | 1 240–1 275 | 1 202 | Proprietary |
| 211 | Mistral Large 2407 | Mistral AI | 1 254 | 1 236–1 271 | 1 230 | Mistral Research |
| 212 | Qwen Max | Alibaba Qwen | 1 254 | 1 228–1 280 | 438 | Qwen |
| 213 | Llama 3.1 405B Instruct · bf16 | Meta | 1 252 | 1 233–1 271 | 894 | Llama 3.1 Community |
| 214 | Athene 70B | Nexusflow | 1 251 | 1 228–1 275 | 585 | CC-BY-NC-4.0 |
| 215 | Llama 3.3 70B Instruct | Meta | 1 251 | 1 234–1 268 | 1 221 | Llama-3.3 |
| 216 | GPT-4 Turbo Preview · 1106-preview | OpenAI | 1 250 | 1 235–1 266 | 2 503 | Proprietary |
| 217 | Llama 3.1 405B Instruct · fp8 | Meta | 1 250 | 1 234–1 266 | 1 587 | Llama 3.1 Community |
| 218 | Gemini 1.5 Pro · 1.5-pro-001 | 1 249 | 1 234–1 264 | 2 677 | Proprietary | |
| 219 | gpt-oss-20b | OpenAI | 1 249 | 1 208–1 289 | 213 | Apache 2.0 |
| 220 | Magistral Medium | Mistral AI | 1 248 | 1 216–1 281 | 361 | Proprietary |
| 221 | GPT-4 Turbo Preview · 0125-preview | OpenAI | 1 245 | 1 230–1 260 | 2 668 | Proprietary |
| 222 | Reka Core | Reka AI | 1 243 | 1 203–1 283 | 183 | Proprietary |
| 223 | Nova Pro 1.0 | Amazon | 1 243 | 1 219–1 266 | 553 | Proprietary |
| 224 | Mistral Large · 2411 | Mistral AI | 1 242 | 1 219–1 264 | 648 | MRL |
| 225 | Claude 3.5 Haiku | Anthropic | 1 236 | 1 221–1 251 | 1 738 | Proprietary |
| 226 | Qwen2.5 72B Instruct | Alibaba Qwen | 1 233 | 1 214–1 252 | 926 | Qwen |
| 227 | Qwen2.5 Plus | Alibaba Qwen | 1 231 | 1 195–1 266 | 257 | Proprietary |
| 228 | Nova Lite 1.0 | Amazon | 1 228 | 1 202–1 255 | 490 | Proprietary |
| 229 | DeepSeek V2.5 · v2.5 | DeepSeek | 1 226 | 1 202–1 251 | 563 | DeepSeek |
| 230 | Phi 4 | Microsoft | 1 222 | 1 197–1 248 | 523 | MIT |
| 231 | Llama 3.1 70B Instruct | Meta | 1 221 | 1 205–1 238 | 1 353 | Llama 3.1 Community |
| 232 | Gemma 2 9B IT SimPO | Princeton NLP | 1 218 | 1 187–1 249 | 318 | MIT |
| 233 | Gemini 1.5 Flash · 001 | 1 216 | 1 201–1 232 | 2 169 | Proprietary | |
| 234 | Command R+ (08-2024) | Cohere | 1 216 | 1 182–1 249 | 267 | CC-BY-NC-4.0 |
| 235 | Mistral Small 3 | Mistral AI | 1 211 | 1 178–1 245 | 301 | Apache 2.0 |
| 236 | Gemma 2 27B | 1 209 | 1 194–1 223 | 2 037 | Gemma license | |
| 237 | Gemini 1.5 Flash-8B | 1 206 | 1 185–1 227 | 798 | Proprietary | |
| 238 | Claude 3 Sonnet | Anthropic | 1 203 | 1 188–1 219 | 3 097 | Proprietary |
| 239 | Jamba 1.5 Large | AI21 Labs | 1 200 | 1 163–1 236 | 234 | Jamba Open |
| 240 | Aya Expanse 32B | Cohere | 1 198 | 1 175–1 222 | 609 | CC-BY-NC-4.0 |
| 241 | GPT-4 · 0314 | OpenAI | 1 198 | 1 179–1 217 | 1 265 | Proprietary |
| 242 | Command R+ | Cohere | 1 195 | 1 180–1 210 | 2 864 | CC-BY-NC-4.0 |
| 243 | Nova Micro 1.0 | Amazon | 1 192 | 1 165–1 219 | 468 | Proprietary |
| 244 | Nemotron-4 340B Instruct | NVIDIA | 1 192 | 1 169–1 215 | 703 | NVIDIA Open Model |
| 245 | Aya Expanse 8B | Cohere | 1 187 | 1 150–1 225 | 240 | CC-BY-NC-4.0 |
| 246 | Gemma 2 9B | 1 186 | 1 170–1 202 | 1 580 | Gemma license | |
| 247 | GLM-4 | Z.ai | 1 184 | 1 154–1 214 | 371 | Proprietary |
| 248 | Reka Flash (2024-09) | Reka AI | 1 178 | 1 140–1 217 | 207 | Proprietary |
| 249 | Mistral Large · 2402 | Mistral AI | 1 178 | 1 161–1 195 | 1 690 | Proprietary |
| 250 | Command R (08-2024) | Cohere | 1 176 | 1 143–1 209 | 275 | CC-BY-NC-4.0 |
| 251 | Claude 3 Haiku | Anthropic | 1 173 | 1 159–1 187 | 3 508 | Proprietary |
| 252 | Llama 3 70B Instruct | Meta | 1 169 | 1 155–1 182 | 4 991 | Llama 3 Community |
| 253 | DeepSeek Coder V2 | DeepSeek | 1 164 | 1 138–1 190 | 516 | DeepSeek License |
| 254 | Jamba 1.5 Mini | AI21 Labs | 1 162 | 1 124–1 200 | 248 | Jamba Open |
| 255 | GPT-4 · 0613 | OpenAI | 1 161 | 1 145–1 177 | 2 215 | Proprietary |
| 256 | Mistral Medium (2023) | Mistral AI | 1 154 | 1 132–1 177 | 852 | Proprietary |
| 257 | Reka Flash 21B · 20240226 | Reka AI | 1 152 | 1 131–1 174 | 798 | Proprietary |
| 258 | Reka Flash 21B · 20240226-online | Reka AI | 1 151 | 1 124–1 177 | 501 | Proprietary |
| 259 | Qwen2 72B Instruct | Alibaba Qwen | 1 151 | 1 133–1 168 | 1 340 | Qianwen LICENSE |
| 260 | Llama 3.1 8B Instruct | Meta | 1 144 | 1 126–1 162 | 1 295 | Llama 3.1 Community |
| 261 | Mixtral 8x22B Instruct | Mistral AI | 1 141 | 1 124–1 158 | 1 653 | Apache 2.0 |
| 262 | Command R | Cohere | 1 127 | 1 109–1 146 | 1 361 | CC-BY-NC-4.0 |
| 263 | Gemini 1.0 Pro | 1 124 | 1 095–1 154 | 450 | Proprietary | |
| 264 | Qwen1.5 110B Chat | Alibaba Qwen | 1 122 | 1 101–1 143 | 928 | Qianwen LICENSE |
| 265 | Gemma 2 2B | 1 114 | 1 095–1 133 | 1 158 | Gemma license | |
| 266 | Mixtral 8x7B Instruct | Mistral AI | 1 114 | 1 097–1 130 | 1 975 | Apache 2.0 |
| 267 | Yi-1.5 34B Chat | 01.AI | 1 111 | 1 090–1 131 | 871 | Apache-2.0 |
| 268 | Llama 3 8B Instruct | Meta | 1 104 | 1 090–1 119 | 3 370 | Llama 3 Community |
| 269 | Phi-3 Medium 4K Instruct | Microsoft | 1 101 | 1 079–1 122 | 851 | MIT |
| 270 | GPT-3.5 Turbo · 0125 | OpenAI | 1 090 | 1 073–1 107 | 1 885 | Proprietary |
| 271 | Qwen1.5 72B Chat | Alibaba Qwen | 1 083 | 1 064–1 103 | 1 091 | Qianwen LICENSE |
| 272 | WizardLM 70B | Microsoft | 1 082 | 1 038–1 127 | 193 | Llama 2 Community |
| 273 | Starling-LM 7B Beta | Berkeley Nest | 1 081 | 1 050–1 112 | 367 | Apache-2.0 |
| 274 | Phi-3 Small 8K Instruct | Microsoft | 1 079 | 1 055–1 104 | 620 | MIT |
| 275 | Snowflake Arctic Instruct | Snowflake | 1 073 | 1 051–1 096 | 947 | Apache 2.0 |
| 276 | InternLM2.5 20B Chat | InternLM | 1 062 | 1 025–1 100 | 265 | Other |
| 277 | OpenChat 3.5 | OpenChat | 1 060 | 1 025–1 094 | 268 | Apache-2.0 |
| 278 | Vicuna 33B | LMSYS | 1 059 | 1 029–1 088 | 428 | Non-commercial |
| 279 | Qwen1.5 32B Chat | Alibaba Qwen | 1 058 | 1 035–1 081 | 777 | Qianwen LICENSE |
| 280 | DBRX Instruct | Databricks | 1 057 | 1 034–1 080 | 742 | DBRX LICENSE |
| 281 | Llama 3.2 3B Instruct | Meta | 1 056 | 1 015–1 096 | 232 | Llama 3.2 |
| 282 | GPT-3.5 Turbo · 1106 | OpenAI | 1 056 | 1 021–1 091 | 325 | Proprietary |
| 283 | Gemma 1.1 7B | 1 053 | 1 032–1 075 | 827 | Gemma license | |
| 284 | Phi-3 Mini 4K Instruct · instruct | Microsoft | 1 044 | 1 020–1 068 | 699 | MIT |
| 285 | Qwen1.5 14B Chat | Alibaba Qwen | 1 042 | 1 016–1 068 | 540 | Qianwen LICENSE |
| 286 | Yi 34B Chat | 01.AI | 1 042 | 1 010–1 074 | 381 | Yi License |
| 287 | Llama 2 70B Chat | Meta | 1 041 | 1 019–1 063 | 878 | Llama 2 Community |
| 288 | Phi-3 Mini 4K Instruct · instruct-june-2024 | Microsoft | 1 031 | 998–1 063 | 397 | MIT |
| 289 | Starling-LM 7B Alpha | Berkeley Nest | 1 025 | 983–1 067 | 199 | CC-BY-NC-4.0 |
| 290 | Llama 3.2 1B Instruct | Meta | 1 013 | 974–1 053 | 247 | Llama 3.2 |
| 291 | Llama 2 13B Chat | Meta | 1 009 | 979–1 039 | 454 | Llama 2 Community |
| 292 | Phi-3 Mini 128K Instruct | Microsoft | 1 006 | 981–1 030 | 727 | MIT |
| 293 | Vicuna 13B | LMSYS | 992 | 956–1 028 | 283 | Llama 2 Community |
| 294 | Zephyr 7B Beta | Hugging Face H4 | 989 | 941–1 037 | 152 | MIT |
| 295 | Mistral 7B Instruct · instruct-v0.2 | Mistral AI | 987 | 960–1 015 | 564 | Apache-2.0 |
| 296 | Llama 2 7B Chat | Meta | 977 | 942–1 012 | 327 | Llama 2 Community |
| 297 | Gemma 1.1 2B | 944 | 909–978 | 365 | Gemma license | |
| 298 | Mistral 7B Instruct · instruct | Mistral AI | 931 | 885–976 | 174 | Apache 2.0 |
| 299 | Qwen1.5 4B Chat | Alibaba Qwen | 902 | 862–942 | 211 | Qianwen LICENSE |
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.