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 «японский», 265 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 | 1 515 | 1 481–1 549 | 365 | Proprietary |
| 2 | Gemini 3 Pro | 1 511 | 1 481–1 541 | 455 | Proprietary | |
| 3 | Claude Fable 5 | Anthropic | 1 507 | 1 477–1 538 | 423 | Proprietary |
| 4 | Qwen3.5 Max | Alibaba Qwen | 1 502 | 1 456–1 549 | 200 | Proprietary |
| 5 | Claude Opus 5 в режиме high | Anthropic | 1 502 | 1 477–1 527 | 715 | Proprietary |
| 6 | GPT-5.5 в режиме high | OpenAI | 1 497 | 1 474–1 519 | 853 | Proprietary |
| 7 | Kimi K3 в режиме max | Moonshot AI | 1 497 | 1 464–1 529 | 373 | Kimi K3 license |
| 8 | Gemini 3 Flash Preview | 1 492 | 1 458–1 527 | 327 | Proprietary | |
| 9 | Gemini 3.1 Pro Preview | 1 490 | 1 471–1 509 | 1 249 | Proprietary | |
| 10 | GPT-5.4 в режиме high | OpenAI | 1 486 | 1 460–1 511 | 643 | Proprietary |
| 11 | Claude Opus 4.6 в режиме high | Anthropic | 1 483 | 1 461–1 506 | 792 | Proprietary |
| 12 | GPT-5.5 · 5.5 | OpenAI | 1 480 | 1 456–1 503 | 776 | Proprietary |
| 13 | GPT-5.6 Sol в режиме xhigh | OpenAI | 1 477 | 1 446–1 508 | 419 | Proprietary |
| 14 | Gemini 3.5 Flash в режиме high | 1 476 | 1 449–1 503 | 533 | Proprietary | |
| 15 | Claude Opus 4.6 | Anthropic | 1 474 | 1 451–1 497 | 825 | Proprietary |
| 16 | Claude Opus 4.7 в режиме high | Anthropic | 1 471 | 1 445–1 496 | 650 | Proprietary |
| 17 | Claude Opus 4.7 | Anthropic | 1 466 | 1 442–1 490 | 752 | Proprietary |
| 18 | Gemini 2.5 Pro | 1 464 | 1 449–1 479 | 2 027 | Proprietary | |
| 19 | Gemini 3.6 Flash в режиме high | 1 463 | 1 434–1 492 | 459 | Proprietary | |
| 20 | Qwen3.8 Max (0902) | Alibaba Qwen | 1 463 | 1 426–1 499 | 300 | Proprietary |
| 21 | Gemini 3.5 Flash в режиме medium | 1 459 | 1 431–1 487 | 539 | Proprietary | |
| 22 | DeepSeek V4 Pro 0423 | DeepSeek | 1 452 | 1 428–1 476 | 691 | MIT |
| 23 | GPT-5.1 в режиме high | OpenAI | 1 451 | 1 423–1 479 | 443 | Proprietary |
| 24 | GPT-5.6 Terra в режиме xhigh | OpenAI | 1 449 | 1 419–1 479 | 440 | Proprietary |
| 25 | GPT-5.4 | OpenAI | 1 448 | 1 423–1 472 | 649 | Proprietary |
| 26 | GLM 5.2 в режиме max | Z.ai | 1 446 | 1 420–1 472 | 580 | MIT |
| 27 | Kimi K2.6 | Moonshot AI | 1 443 | 1 414–1 472 | 467 | Modified MIT |
| 28 | Grok 4.20 · beta1 | xAI | 1 442 | 1 401–1 483 | 216 | Proprietary |
| 29 | Claude Opus 4.8 в режиме high | Anthropic | 1 439 | 1 416–1 463 | 754 | Proprietary |
| 30 | Claude Opus 4.8 | Anthropic | 1 435 | 1 411–1 459 | 720 | Proprietary |
| 31 | GLM 5.3 в режиме max | Z.ai | 1 433 | 1 391–1 475 | 209 | MIT |
| 32 | GLM 5.1 | Z.ai | 1 431 | 1 406–1 456 | 642 | MIT |
| 33 | Qwen3.5 397B A17B | Alibaba Qwen | 1 429 | 1 408–1 451 | 880 | Apache 2.0 |
| 34 | Grok 4.5 | xAI | 1 429 | 1 398–1 460 | 406 | Proprietary |
| 35 | Gemini 3 Flash Preview в режиме minimal | 1 428 | 1 405–1 451 | 803 | Proprietary | |
| 36 | Muse Spark 1.1 | Meta | 1 428 | 1 399–1 457 | 466 | Proprietary |
| 37 | ERNIE 5.1 | Baidu | 1 426 | 1 398–1 454 | 493 | Proprietary |
| 38 | Claude Sonnet 5 в режиме high | Anthropic | 1 426 | 1 399–1 453 | 547 | Proprietary |
| 39 | GPT-5.6 Luna в режиме xhigh | OpenAI | 1 423 | 1 393–1 454 | 432 | Proprietary |
| 40 | Gemini 3.5 Flash Lite | 1 422 | 1 392–1 452 | 441 | Proprietary | |
| 41 | GLM 5.3 Flash | Z.ai | 1 420 | 1 375–1 465 | 182 | MIT |
| 42 | GPT-5.2 в режиме high | OpenAI | 1 420 | 1 390–1 450 | 426 | Proprietary |
| 43 | GPT-5.2 Chat | OpenAI | 1 420 | 1 383–1 457 | 256 | Proprietary |
| 44 | Kimi K2.5 | Moonshot AI | 1 420 | 1 398–1 442 | 803 | Modified MIT |
| 45 | GPT-5.1 | OpenAI | 1 418 | 1 389–1 447 | 433 | Proprietary |
| 46 | Claude Opus 4.5 в режиме high | Anthropic | 1 418 | 1 386–1 450 | 361 | Proprietary |
| 47 | Claude Sonnet 4.6 | Anthropic | 1 417 | 1 394–1 441 | 754 | Proprietary |
| 48 | GLM 4.5 | Z.ai | 1 416 | 1 393–1 439 | 765 | MIT |
| 49 | Qwen3.7 Plus | Alibaba Qwen | 1 416 | 1 391–1 441 | 642 | Proprietary |
| 50 | GPT-4.5 Preview | OpenAI | 1 416 | 1 378–1 454 | 271 | Proprietary |
| 51 | MiMo-V2.5-Pro | Xiaomi | 1 414 | 1 391–1 437 | 806 | MIT |
| 52 | Claude Opus 4.5 | Anthropic | 1 413 | 1 388–1 438 | 616 | Proprietary |
| 53 | Inkling | Thinking Machines Lab | 1 412 | 1 382–1 443 | 408 | Apache 2.0 |
| 54 | DeepSeek V4 Pro 0423 в режиме high | DeepSeek | 1 411 | 1 387–1 436 | 646 | MIT |
| 55 | GPT-5 | OpenAI | 1 411 | 1 384–1 438 | 517 | Proprietary |
| 56 | Gemini 3.1 Flash Lite Preview | 1 410 | 1 385–1 435 | 609 | Proprietary | |
| 57 | GLM 5 | Z.ai | 1 410 | 1 373–1 447 | 262 | MIT |
| 58 | Grok 4.20 · beta-0309-reasoning | xAI | 1 408 | 1 384–1 432 | 673 | Proprietary |
| 59 | Grok 4.20 Multi-Agent | xAI | 1 407 | 1 382–1 432 | 585 | Proprietary |
| 60 | GPT-5.5 · 5.5-instant | OpenAI | 1 405 | 1 367–1 443 | 258 | Proprietary |
| 61 | ChatGPT-4o (latest) | OpenAI | 1 405 | 1 387–1 423 | 1 261 | Proprietary |
| 62 | o3 | OpenAI | 1 403 | 1 386–1 421 | 1 404 | Proprietary |
| 63 | Gemini 2.5 Flash · flash-preview-09-2025 | 1 403 | 1 371–1 435 | 347 | Proprietary | |
| 64 | GPT-5 в режиме high | OpenAI | 1 400 | 1 379–1 422 | 892 | Proprietary |
| 65 | Qwen3 Max | Alibaba Qwen | 1 400 | 1 372–1 429 | 435 | Proprietary |
| 66 | Qwen3 235B A22B Instruct 2507 | Alibaba Qwen | 1 400 | 1 384–1 417 | 1 400 | Apache 2.0 |
| 67 | DeepSeek V4 Flash 0423 в режиме high | DeepSeek | 1 399 | 1 374–1 425 | 598 | MIT |
| 68 | Seed 2.0 Pro | ByteDance Seed | 1 399 | 1 376–1 422 | 812 | Proprietary |
| 69 | Qwen3 Next 80B A3B Instruct | Alibaba Qwen | 1 398 | 1 365–1 432 | 310 | Apache 2.0 |
| 70 | Gemini 2.5 Flash · flash | 1 397 | 1 383–1 411 | 1 993 | Proprietary | |
| 71 | Mistral Large 3 2512 | Mistral AI | 1 397 | 1 374–1 421 | 720 | Apache 2.0 |
| 72 | Grok 4 | xAI | 1 397 | 1 377–1 418 | 904 | Proprietary |
| 73 | Inkling Small | Thinking Machines Lab | 1 396 | 1 361–1 431 | 316 | Apache 2.0 |
| 74 | R1 0528 | DeepSeek | 1 393 | 1 368–1 419 | 607 | MIT |
| 75 | GLM 4.6 | Z.ai | 1 393 | 1 362–1 424 | 353 | MIT |
| 76 | GPT-5.2 | OpenAI | 1 392 | 1 369–1 415 | 747 | Proprietary |
| 77 | Grok 4.1 · 4.1 | xAI | 1 391 | 1 369–1 414 | 685 | Proprietary |
| 78 | Claude Sonnet 4.5 | Anthropic | 1 390 | 1 368–1 413 | 749 | Proprietary |
| 79 | Grok 3 | xAI | 1 389 | 1 368–1 410 | 823 | Proprietary |
| 80 | Qwen3 235B A22B Thinking 2507 | Alibaba Qwen | 1 387 | 1 358–1 417 | 432 | Apache 2.0 |
| 81 | Qwen3.6 Plus | Alibaba Qwen | 1 387 | 1 360–1 415 | 523 | Proprietary |
| 82 | DeepSeek V4 Flash 0423 | DeepSeek | 1 386 | 1 360–1 411 | 632 | MIT |
| 83 | Grok 4.6 в режиме high | xAI | 1 385 | 1 350–1 421 | 288 | Proprietary |
| 84 | Qwen3 VL 235B A22B Instruct | Alibaba Qwen | 1 384 | 1 339–1 430 | 169 | Apache 2.0 |
| 85 | MiniMax M3 | MiniMax | 1 384 | 1 359–1 408 | 718 | MiniMax Community License |
| 86 | ERNIE 5.0 | Baidu | 1 382 | 1 350–1 413 | 372 | Proprietary |
| 87 | Grok 4.1 · 4.1-thinking | xAI | 1 381 | 1 357–1 405 | 618 | Proprietary |
| 88 | Claude Opus 4.1 · 20250805-thinking-16k | Anthropic | 1 378 | 1 356–1 401 | 654 | Proprietary |
| 89 | Amazon Nova Experimental Chat · 11-10 | Amazon | 1 378 | 1 336–1 420 | 197 | Proprietary |
| 90 | Mistral Medium 3.1 | Mistral AI | 1 378 | 1 360–1 396 | 1 100 | Proprietary |
| 91 | Grok 4.3 | xAI | 1 377 | 1 354–1 400 | 819 | Proprietary |
| 92 | DeepSeek V3.2 · v3.2-thinking | DeepSeek | 1 376 | 1 346–1 406 | 362 | MIT |
| 93 | Claude Opus 4.1 · 20250805 | Anthropic | 1 376 | 1 358–1 394 | 1 169 | Proprietary |
| 94 | MiMo-V2 Pro | Xiaomi | 1 375 | 1 331–1 418 | 167 | Proprietary |
| 95 | Claude Sonnet 4.5 в режиме high | Anthropic | 1 375 | 1 353–1 397 | 764 | Proprietary |
| 96 | GPT-5.4 Mini в режиме high | OpenAI | 1 375 | 1 349–1 401 | 587 | Proprietary |
| 97 | DeepSeek V3.1 · v3.1 | DeepSeek | 1 374 | 1 343–1 405 | 329 | MIT |
| 98 | Qwen3 235B A22B · a22b-no-thinking | Alibaba Qwen | 1 371 | 1 351–1 392 | 877 | Apache 2.0 |
| 99 | Qwen3.8 27B | Alibaba Qwen | 1 370 | 1 329–1 412 | 221 | Apache 2.0 |
| 100 | LongCat-Flash-Chat · chat-2602-exp | Meituan | 1 370 | 1 331–1 409 | 251 | Proprietary |
| 101 | Qwen3.5-Flash | Alibaba Qwen | 1 366 | 1 341–1 391 | 622 | Proprietary |
| 102 | Qwen3.5-122B-A10B | Alibaba Qwen | 1 364 | 1 327–1 402 | 243 | Apache 2.0 |
| 103 | Amazon Nova Experimental Chat · 10-20 | Amazon | 1 364 | 1 313–1 415 | 134 | Proprietary |
| 104 | DeepSeek V3.1 · v3.1-thinking | DeepSeek | 1 362 | 1 326–1 398 | 255 | MIT |
| 105 | Gemini 2.5 Flash Lite · 09-2025-no-thinking | 1 360 | 1 334–1 387 | 482 | Proprietary | |
| 106 | DeepSeek V3.2 · v3.2 | DeepSeek | 1 359 | 1 330–1 388 | 441 | MIT |
| 107 | Grok 4 Fast | xAI | 1 359 | 1 317–1 400 | 207 | Proprietary |
| 108 | Kimi K2 Thinking | Moonshot AI | 1 353 | 1 327–1 379 | 547 | Modified MIT |
| 109 | GPT-5.3 Chat | OpenAI | 1 351 | 1 312–1 391 | 264 | Proprietary |
| 110 | Step 3.5 Flash | StepFun | 1 351 | 1 326–1 376 | 650 | Apache 2.0 |
| 111 | Kimi K2 0905 | Moonshot AI | 1 349 | 1 309–1 389 | 210 | Modified MIT |
| 112 | Qwen3.5-27B | Alibaba Qwen | 1 348 | 1 307–1 389 | 218 | Apache 2.0 |
| 113 | GLM 4.5 Air | Z.ai | 1 348 | 1 327–1 369 | 831 | MIT |
| 114 | Gemini 2.5 Flash Lite · 06-17-thinking | 1 348 | 1 327–1 369 | 796 | Proprietary | |
| 115 | o1 · 2024-12-17 | OpenAI | 1 346 | 1 320–1 372 | 581 | Proprietary |
| 116 | Grok 3 Mini | xAI | 1 343 | 1 318–1 368 | 580 | Proprietary |
| 117 | Grok 4.1 Fast | xAI | 1 342 | 1 312–1 371 | 421 | Proprietary |
| 118 | GPT-5 Mini в режиме high | OpenAI | 1 341 | 1 313–1 370 | 448 | Proprietary |
| 119 | Qwen3 30B A3B Instruct 2507 | Alibaba Qwen | 1 339 | 1 317–1 361 | 755 | Apache 2.0 |
| 120 | Hunyuan TurboS | Tencent | 1 338 | 1 305–1 371 | 329 | Proprietary |
| 121 | Claude Haiku 4.5 | Anthropic | 1 338 | 1 320–1 356 | 1 438 | Proprietary |
| 122 | GPT-5.4 Nano в режиме high | OpenAI | 1 337 | 1 310–1 363 | 582 | Proprietary |
| 123 | Kimi K2 0711 | Moonshot AI | 1 337 | 1 316–1 357 | 884 | Modified MIT |
| 124 | DeepSeek V3 0324 | DeepSeek | 1 334 | 1 315–1 353 | 1 044 | MIT |
| 125 | Claude Opus 4 · 20250514-thinking-16k | Anthropic | 1 333 | 1 311–1 355 | 820 | Proprietary |
| 126 | gpt-oss-120b | OpenAI | 1 332 | 1 307–1 358 | 568 | Apache 2.0 |
| 127 | Claude Opus 4 · 20250514 | Anthropic | 1 331 | 1 312–1 350 | 1 099 | Proprietary |
| 128 | DeepSeek V3.2 Exp | DeepSeek | 1 331 | 1 280–1 381 | 140 | MIT |
| 129 | LongCat-Flash-Chat · chat | Meituan | 1 329 | 1 288–1 370 | 186 | MIT |
| 130 | Qwen3.5-35B-A3B | Alibaba Qwen | 1 329 | 1 289–1 368 | 244 | Apache 2.0 |
| 131 | MiMo-V2 Flash | Xiaomi | 1 326 | 1 296–1 357 | 422 | MIT |
| 132 | R1 | DeepSeek | 1 324 | 1 299–1 350 | 468 | MIT |
| 133 | GPT-4.1 | OpenAI | 1 320 | 1 301–1 339 | 1 115 | Proprietary |
| 134 | Mistral Medium 3 | Mistral AI | 1 317 | 1 297–1 338 | 892 | Proprietary |
| 135 | MiMo-V2 Omni | Xiaomi | 1 316 | 1 279–1 352 | 271 | Proprietary |
| 136 | Grok 3 Mini в режиме high | xAI | 1 315 | 1 285–1 345 | 389 | Proprietary |
| 137 | Trinity Large Thinking | Arcee AI | 1 314 | 1 276–1 352 | 265 | Apache 2.0 |
| 138 | Qwen3 235B A22B · a22b | Alibaba Qwen | 1 314 | 1 290–1 337 | 701 | Apache 2.0 |
| 139 | o4 Mini | OpenAI | 1 309 | 1 289–1 328 | 1 059 | Proprietary |
| 140 | Qwen3 Coder 480B A35B | Alibaba Qwen | 1 308 | 1 288–1 329 | 902 | Apache 2.0 |
| 141 | MiMo-V2.5 | Xiaomi | 1 304 | 1 275–1 332 | 509 | MIT |
| 142 | Gemini 2.0 Flash-Lite | 1 302 | 1 277–1 326 | 542 | Proprietary | |
| 143 | Llama 3.3 Nemotron Super 49B v1.5 | NVIDIA | 1 301 | 1 272–1 331 | 416 | Nvidia Open |
| 144 | Claude Sonnet 4 · 20250514 | Anthropic | 1 301 | 1 281–1 321 | 951 | Proprietary |
| 145 | Qwen2.5 Max | Alibaba Qwen | 1 299 | 1 277–1 322 | 727 | Proprietary |
| 146 | o1 · preview | OpenAI | 1 294 | 1 270–1 318 | 674 | Proprietary |
| 147 | Gemini 2.0 Flash | 1 293 | 1 274–1 312 | 1 007 | Proprietary | |
| 148 | Gemini 1.5 Pro · 1.5-pro-002 | 1 292 | 1 275–1 310 | 1 210 | Proprietary | |
| 149 | GPT-4.1 Mini | OpenAI | 1 291 | 1 272–1 311 | 987 | Proprietary |
| 150 | Trinity Large | Arcee AI | 1 289 | 1 248–1 330 | 212 | Apache 2.0 |
| 151 | MiniMax M2.1 | MiniMax | 1 287 | 1 240–1 335 | 170 | MIT |
| 152 | Gemma 3 27B | 1 287 | 1 268–1 306 | 1 041 | Gemma | |
| 153 | Claude Sonnet 4 · 20250514-thinking-32k | Anthropic | 1 287 | 1 264–1 309 | 799 | Proprietary |
| 154 | o3 Mini High | OpenAI | 1 286 | 1 256–1 316 | 374 | Proprietary |
| 155 | Command A | Cohere | 1 284 | 1 266–1 302 | 1 227 | CC-BY-NC-4.0 |
| 156 | Qwen3 Next 80B A3B Thinking | Alibaba Qwen | 1 282 | 1 241–1 324 | 201 | Apache 2.0 |
| 157 | Nova 2 Lite | Amazon | 1 276 | 1 217–1 336 | 106 | Proprietary |
| 158 | Mistral Small 3.2 24B | Mistral AI | 1 276 | 1 250–1 302 | 511 | Apache 2.0 |
| 159 | Gemma 3n E4B | 1 274 | 1 252–1 297 | 722 | Gemma | |
| 160 | Claude 3.7 Sonnet · 20250219-thinking-32k | Anthropic | 1 268 | 1 248–1 288 | 1 001 | Proprietary |
| 161 | Nemotron 3 Nano 30B A3B | NVIDIA | 1 267 | 1 214–1 320 | 139 | NVIDIA Open Model |
| 162 | DeepSeek V3 | DeepSeek | 1 266 | 1 243–1 290 | 534 | DeepSeek |
| 163 | o3 Mini | OpenAI | 1 265 | 1 248–1 281 | 1 395 | Proprietary |
| 164 | MiniMax M2.7 | MiniMax | 1 264 | 1 240–1 287 | 867 | Modified MIT |
| 165 | QwQ 32B | Alibaba Qwen | 1 264 | 1 239–1 289 | 608 | Apache 2.0 |
| 166 | Claude 3.7 Sonnet · 20250219 | Anthropic | 1 261 | 1 241–1 280 | 990 | Proprietary |
| 167 | GPT-4o (2024-05-13) | OpenAI | 1 257 | 1 241–1 272 | 2 674 | Proprietary |
| 168 | Qwen3 30B A3B | Alibaba Qwen | 1 256 | 1 233–1 278 | 723 | Apache 2.0 |
| 169 | Gemini 1.5 Flash · 002 | 1 252 | 1 230–1 274 | 753 | Proprietary | |
| 170 | Qwen-Plus | Alibaba Qwen | 1 251 | 1 212–1 290 | 206 | Proprietary |
| 171 | gpt-oss-20b | OpenAI | 1 245 | 1 206–1 284 | 241 | Apache 2.0 |
| 172 | o1-mini | OpenAI | 1 245 | 1 226–1 263 | 1 093 | Proprietary |
| 173 | Grok 2 | xAI | 1 244 | 1 227–1 261 | 1 478 | Proprietary |
| 174 | Gemini 1.5 Pro · advanced-0514 | 1 242 | 1 221–1 263 | 1 212 | Proprietary | |
| 175 | Gemini 1.5 Pro · 1.5-pro-001 | 1 239 | 1 222–1 256 | 1 935 | Proprietary | |
| 176 | Claude 3.5 Sonnet · 20241022 | Anthropic | 1 234 | 1 220–1 248 | 2 032 | Proprietary |
| 177 | GPT-4o (2024-08-06) | OpenAI | 1 234 | 1 215–1 253 | 1 104 | Proprietary |
| 178 | Claude 3.5 Sonnet · 20240620 | Anthropic | 1 231 | 1 215–1 247 | 2 127 | Proprietary |
| 179 | Yi-Lightning | 01.AI | 1 228 | 1 204–1 252 | 565 | Proprietary |
| 180 | DeepSeek V2.5 · v2.5-1210 | DeepSeek | 1 228 | 1 186–1 269 | 161 | DeepSeek |
| 181 | GLM-4-Plus · plus-0111 | Z.ai | 1 225 | 1 187–1 263 | 201 | Proprietary |
| 182 | MiniMax M1 | MiniMax | 1 220 | 1 198–1 242 | 826 | Apache 2.0 |
| 183 | GPT-4o-mini (2024-07-18) | OpenAI | 1 216 | 1 200–1 231 | 1 793 | Proprietary |
| 184 | Athene V2 Chat | Nexusflow | 1 215 | 1 192–1 239 | 584 | NexusFlow |
| 185 | Grok 2 Mini | xAI | 1 213 | 1 195–1 231 | 1 225 | Proprietary |
| 186 | Mistral Small 3.1 24B | Mistral AI | 1 210 | 1 188–1 232 | 796 | Apache 2.0 |
| 187 | Llama 4 Scout | Meta | 1 209 | 1 187–1 232 | 751 | Llama |
| 188 | GPT-4 Turbo Preview · 1106-preview | OpenAI | 1 209 | 1 191–1 228 | 1 422 | Proprietary |
| 189 | Llama 3.1 405B Instruct · bf16 | Meta | 1 208 | 1 187–1 228 | 846 | Llama 3.1 Community |
| 190 | GLM-4-Plus · plus | Z.ai | 1 207 | 1 180–1 235 | 485 | Proprietary |
| 191 | Llama 4 Maverick | Meta | 1 207 | 1 185–1 228 | 931 | Llama 4 |
| 192 | Qwen2.5 Plus | Alibaba Qwen | 1 207 | 1 174–1 239 | 300 | Proprietary |
| 193 | Qwen Max | Alibaba Qwen | 1 205 | 1 173–1 237 | 312 | Qwen |
| 194 | Claude 3 Opus | Anthropic | 1 204 | 1 190–1 218 | 3 650 | Proprietary |
| 195 | Nova Pro 1.0 | Amazon | 1 200 | 1 174–1 226 | 511 | Proprietary |
| 196 | GPT-4.1 Nano | OpenAI | 1 199 | 1 154–1 244 | 160 | Proprietary |
| 197 | GPT-4 Turbo Preview · 0125-preview | OpenAI | 1 195 | 1 177–1 214 | 1 474 | Proprietary |
| 198 | GPT-4 Turbo | OpenAI | 1 194 | 1 177–1 210 | 2 112 | Proprietary |
| 199 | DeepSeek V2.5 · v2.5 | DeepSeek | 1 190 | 1 163–1 217 | 479 | DeepSeek |
| 200 | Mistral Large 2407 | Mistral AI | 1 187 | 1 170–1 205 | 1 249 | Mistral Research |
| 201 | Gemini 1.5 Flash · 001 | 1 186 | 1 168–1 204 | 1 522 | Proprietary | |
| 202 | Qwen2.5 72B Instruct | Alibaba Qwen | 1 180 | 1 159–1 201 | 788 | Qwen |
| 203 | Gemma 2 27B | 1 175 | 1 160–1 191 | 1 808 | Gemma license | |
| 204 | Claude 3.5 Haiku | Anthropic | 1 175 | 1 159–1 191 | 1 527 | Proprietary |
| 205 | Magistral Medium | Mistral AI | 1 174 | 1 139–1 210 | 340 | Proprietary |
| 206 | Llama 3.1 405B Instruct · fp8 | Meta | 1 171 | 1 155–1 188 | 1 606 | Llama 3.1 Community |
| 207 | Mistral Large · 2411 | Mistral AI | 1 170 | 1 147–1 193 | 629 | MRL |
| 208 | MiniMax M2.5 | MiniMax | 1 168 | 1 130–1 207 | 357 | Modified MIT |
| 209 | Command R+ (08-2024) | Cohere | 1 166 | 1 128–1 204 | 210 | CC-BY-NC-4.0 |
| 210 | Aya Expanse 32B | Cohere | 1 163 | 1 140–1 186 | 670 | CC-BY-NC-4.0 |
| 211 | Phi 4 | Microsoft | 1 158 | 1 132–1 185 | 500 | MIT |
| 212 | Command R+ | Cohere | 1 155 | 1 135–1 174 | 1 487 | CC-BY-NC-4.0 |
| 213 | Nova Micro 1.0 | Amazon | 1 154 | 1 127–1 182 | 443 | Proprietary |
| 214 | Nova Lite 1.0 | Amazon | 1 153 | 1 125–1 181 | 403 | Proprietary |
| 215 | Gemini 1.5 Flash-8B | 1 150 | 1 127–1 173 | 691 | Proprietary | |
| 216 | Llama 3.3 70B Instruct | Meta | 1 149 | 1 130–1 167 | 1 157 | Llama-3.3 |
| 217 | Gemma 2 9B | 1 144 | 1 127–1 162 | 1 301 | Gemma license | |
| 218 | Command R (08-2024) | Cohere | 1 143 | 1 110–1 176 | 254 | CC-BY-NC-4.0 |
| 219 | Gemma 2 9B IT SimPO | Princeton NLP | 1 140 | 1 107–1 172 | 291 | MIT |
| 220 | Athene 70B | Nexusflow | 1 138 | 1 115–1 161 | 676 | CC-BY-NC-4.0 |
| 221 | GPT-4 · 0314 | OpenAI | 1 136 | 1 111–1 162 | 619 | Proprietary |
| 222 | Llama 3.1 70B Instruct | Meta | 1 132 | 1 115–1 149 | 1 428 | Llama 3.1 Community |
| 223 | Claude 3 Sonnet | Anthropic | 1 131 | 1 113–1 149 | 1 753 | Proprietary |
| 224 | Nemotron-4 340B Instruct | NVIDIA | 1 128 | 1 099–1 157 | 434 | NVIDIA Open Model |
| 225 | DeepSeek Coder V2 | DeepSeek | 1 126 | 1 094–1 158 | 345 | DeepSeek License |
| 226 | Aya Expanse 8B | Cohere | 1 120 | 1 085–1 156 | 231 | CC-BY-NC-4.0 |
| 227 | Jamba 1.5 Large | AI21 Labs | 1 118 | 1 082–1 155 | 242 | Jamba Open |
| 228 | GPT-4 · 0613 | OpenAI | 1 114 | 1 094–1 134 | 1 241 | Proprietary |
| 229 | Mistral Small 3 | Mistral AI | 1 111 | 1 075–1 147 | 269 | Apache 2.0 |
| 230 | Qwen2 72B Instruct | Alibaba Qwen | 1 111 | 1 090–1 132 | 901 | Qianwen LICENSE |
| 231 | Claude 3 Haiku | Anthropic | 1 102 | 1 086–1 119 | 2 063 | Proprietary |
| 232 | Command R | Cohere | 1 094 | 1 071–1 117 | 856 | CC-BY-NC-4.0 |
| 233 | Gemma 2 2B | 1 083 | 1 065–1 102 | 1 228 | Gemma license | |
| 234 | Reka Flash 21B · 20240226-online | Reka AI | 1 075 | 1 040–1 110 | 274 | Proprietary |
| 235 | Qwen1.5 110B Chat | Alibaba Qwen | 1 074 | 1 047–1 101 | 599 | Qianwen LICENSE |
| 236 | Qwen1.5 72B Chat | Alibaba Qwen | 1 061 | 1 033–1 089 | 495 | Qianwen LICENSE |
| 237 | Llama 3.1 8B Instruct | Meta | 1 061 | 1 042–1 080 | 1 234 | Llama 3.1 Community |
| 238 | Reka Flash 21B · 20240226 | Reka AI | 1 049 | 1 020–1 078 | 440 | Proprietary |
| 239 | GPT-3.5 Turbo | OpenAI | 1 043 | 1 021–1 065 | 965 | Proprietary |
| 240 | Phi-3 Medium 4K Instruct | Microsoft | 1 042 | 1 018–1 066 | 683 | MIT |
| 241 | Mixtral 8x22B Instruct | Mistral AI | 1 037 | 1 015–1 058 | 972 | Apache 2.0 |
| 242 | Qwen1.5 32B Chat | Alibaba Qwen | 1 027 | 993–1 061 | 324 | Qianwen LICENSE |
| 243 | Gemini 1.0 Pro | 1 023 | 978–1 067 | 173 | Proprietary | |
| 244 | Yi-1.5 34B Chat | 01.AI | 1 021 | 993–1 049 | 506 | Apache-2.0 |
| 245 | Qwen1.5 14B Chat | Alibaba Qwen | 1 019 | 979–1 059 | 217 | Qianwen LICENSE |
| 246 | Jamba 1.5 Mini | AI21 Labs | 1 019 | 983–1 055 | 270 | Jamba Open |
| 247 | Llama 3 70B Instruct | Meta | 1 016 | 1 001–1 032 | 3 276 | Llama 3 Community |
| 248 | Mistral Medium (2023) | Mistral AI | 1 012 | 979–1 046 | 337 | Proprietary |
| 249 | Mistral Large · 2402 | Mistral AI | 1 008 | 985–1 031 | 945 | Proprietary |
| 250 | Starling-LM 7B Beta | Berkeley Nest | 1 001 | 957–1 044 | 181 | Apache-2.0 |
| 251 | Yi 34B Chat | 01.AI | 993 | 949–1 037 | 197 | Yi License |
| 252 | DBRX Instruct | Databricks | 990 | 959–1 020 | 426 | DBRX LICENSE |
| 253 | Gemma 1.1 7B | 971 | 940–1 001 | 449 | Gemma license | |
| 254 | Snowflake Arctic Instruct | Snowflake | 969 | 944–995 | 667 | Apache 2.0 |
| 255 | Llama 3 8B Instruct | Meta | 967 | 949–985 | 1 992 | Llama 3 Community |
| 256 | Phi-3 Small 8K Instruct | Microsoft | 966 | 934–997 | 390 | MIT |
| 257 | Vicuna 33B | LMSYS | 959 | 917–1 000 | 195 | Non-commercial |
| 258 | Phi-3 Mini 4K Instruct · instruct-june-2024 | Microsoft | 954 | 922–987 | 354 | MIT |
| 259 | Phi-3 Mini 4K Instruct · instruct | Microsoft | 935 | 901–969 | 384 | MIT |
| 260 | Mixtral 8x7B Instruct | Mistral AI | 931 | 909–954 | 960 | Apache 2.0 |
| 261 | Llama 2 70B Chat | Meta | 927 | 893–960 | 341 | Llama 2 Community |
| 262 | Vicuna 13B | LMSYS | 918 | 866–970 | 125 | Llama 2 Community |
| 263 | Phi-3 Mini 128K Instruct | Microsoft | 899 | 867–931 | 385 | MIT |
| 264 | Llama 2 13B Chat | Meta | 894 | 844–943 | 175 | Llama 2 Community |
| 265 | Mistral 7B Instruct | Mistral AI | 878 | 832–924 | 167 | 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.