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 «корейский», 269 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 | 1 532 | 1 503–1 561 | 424 | Proprietary |
| 2 | Claude Opus 5 в режиме high | Anthropic | 1 506 | 1 484–1 529 | 800 | Proprietary |
| 3 | Claude Fable 5 | Anthropic | 1 491 | 1 465–1 517 | 600 | Proprietary |
| 4 | GLM 5.3 в режиме max | Z.ai | 1 478 | 1 435–1 520 | 183 | MIT |
| 5 | Qwen3.8 Max (0902) | Alibaba Qwen | 1 469 | 1 437–1 500 | 368 | Proprietary |
| 6 | Kimi K3 в режиме max | Moonshot AI | 1 467 | 1 440–1 494 | 483 | Kimi K3 license |
| 7 | GPT-5.5 в режиме high | OpenAI | 1 462 | 1 443–1 481 | 1 151 | Proprietary |
| 8 | Claude Opus 4.6 | Anthropic | 1 461 | 1 444–1 479 | 1 418 | Proprietary |
| 9 | Claude Opus 4.7 | Anthropic | 1 461 | 1 442–1 481 | 1 096 | Proprietary |
| 10 | Gemini 3.1 Pro Preview | 1 460 | 1 444–1 475 | 1 955 | Proprietary | |
| 11 | Muse Spark | Meta | 1 460 | 1 421–1 499 | 248 | Proprietary |
| 12 | Muse Spark 1.1 | Meta | 1 459 | 1 433–1 484 | 563 | Proprietary |
| 13 | DeepSeek V4 Pro 0813 в режиме high | DeepSeek | 1 456 | 1 413–1 499 | 166 | MIT |
| 14 | Claude Opus 4.7 в режиме high | Anthropic | 1 456 | 1 436–1 476 | 1 045 | Proprietary |
| 15 | Gemini 3.5 Flash в режиме medium | 1 454 | 1 431–1 477 | 696 | Proprietary | |
| 16 | GPT-5.4 в режиме high | OpenAI | 1 452 | 1 433–1 472 | 1 076 | Proprietary |
| 17 | Gemini 3 Pro | 1 450 | 1 426–1 473 | 704 | Proprietary | |
| 18 | Gemini 3.5 Flash в режиме high | 1 448 | 1 426–1 470 | 786 | Proprietary | |
| 19 | Claude Opus 4.6 в режиме high | Anthropic | 1 448 | 1 430–1 466 | 1 272 | Proprietary |
| 20 | Gemini 3 Flash Preview | 1 445 | 1 419–1 470 | 575 | Proprietary | |
| 21 | GPT-5.5 · 5.5 | OpenAI | 1 445 | 1 425–1 464 | 1 139 | Proprietary |
| 22 | GLM 5.2 в режиме max | Z.ai | 1 440 | 1 417–1 462 | 740 | MIT |
| 23 | GPT-5.4 | OpenAI | 1 438 | 1 419–1 457 | 1 134 | Proprietary |
| 24 | Gemini 2.5 Pro | 1 435 | 1 422–1 448 | 2 469 | Proprietary | |
| 25 | MiMo-V2.5-Pro | Xiaomi | 1 435 | 1 416–1 454 | 1 170 | MIT |
| 26 | Qwen3.5 Max | Alibaba Qwen | 1 432 | 1 403–1 462 | 398 | Proprietary |
| 27 | Gemini 3.6 Flash в режиме high | 1 432 | 1 405–1 459 | 503 | Proprietary | |
| 28 | Claude Opus 4.8 в режиме high | Anthropic | 1 432 | 1 411–1 453 | 931 | Proprietary |
| 29 | ERNIE 5.1 | Baidu | 1 430 | 1 405–1 455 | 640 | Proprietary |
| 30 | Claude Opus 4.8 | Anthropic | 1 428 | 1 407–1 449 | 900 | Proprietary |
| 31 | Kimi K2.6 | Moonshot AI | 1 428 | 1 405–1 451 | 713 | Modified MIT |
| 32 | GPT-5.6 Sol в режиме xhigh | OpenAI | 1 428 | 1 401–1 454 | 547 | Proprietary |
| 33 | GLM 5 | Z.ai | 1 424 | 1 398–1 449 | 538 | MIT |
| 34 | Claude Opus 4.5 | Anthropic | 1 423 | 1 405–1 442 | 1 133 | Proprietary |
| 35 | GLM 5.1 | Z.ai | 1 423 | 1 403–1 443 | 972 | MIT |
| 36 | DeepSeek V4 Pro 0423 в режиме high | DeepSeek | 1 421 | 1 401–1 441 | 959 | MIT |
| 37 | GPT-5.6 Terra в режиме xhigh | OpenAI | 1 419 | 1 394–1 444 | 579 | Proprietary |
| 38 | Grok 4.20 Multi-Agent | xAI | 1 418 | 1 398–1 438 | 1 012 | Proprietary |
| 39 | Gemini 3 Flash Preview в режиме minimal | 1 418 | 1 401–1 435 | 1 383 | Proprietary | |
| 40 | Qwen3.7 Plus | Alibaba Qwen | 1 416 | 1 395–1 438 | 777 | Proprietary |
| 41 | Grok 4.20 · beta1 | xAI | 1 416 | 1 387–1 444 | 457 | Proprietary |
| 42 | Grok 4.20 · beta-0309-reasoning | xAI | 1 415 | 1 395–1 435 | 1 030 | Proprietary |
| 43 | Seed 2.0 Pro | ByteDance Seed | 1 413 | 1 395–1 431 | 1 295 | Proprietary |
| 44 | Kimi K2.5 · thinking | Moonshot AI | 1 412 | 1 395–1 430 | 1 330 | Modified MIT |
| 45 | Claude Sonnet 4.6 | Anthropic | 1 412 | 1 393–1 431 | 1 108 | Proprietary |
| 46 | GPT-5.6 Luna в режиме xhigh | OpenAI | 1 411 | 1 386–1 436 | 568 | Proprietary |
| 47 | DeepSeek V4 Pro 0423 | DeepSeek | 1 410 | 1 391–1 430 | 1 069 | MIT |
| 48 | Claude Sonnet 5 в режиме high | Anthropic | 1 407 | 1 383–1 431 | 649 | Proprietary |
| 49 | Grok 4.1 · 4.1 | xAI | 1 407 | 1 388–1 426 | 1 127 | Proprietary |
| 50 | Gemini 3.5 Flash Lite | 1 406 | 1 380–1 432 | 523 | Proprietary | |
| 51 | ERNIE 5.0 · 0110 | Baidu | 1 404 | 1 381–1 428 | 632 | Proprietary |
| 52 | Hunyuan T1 | Tencent | 1 403 | 1 358–1 449 | 180 | Proprietary |
| 53 | Qwen3.8 27B | Alibaba Qwen | 1 402 | 1 360–1 445 | 182 | Apache 2.0 |
| 54 | Inkling | Thinking Machines Lab | 1 402 | 1 375–1 429 | 500 | Apache 2.0 |
| 55 | Grok 4.6 в режиме high | xAI | 1 402 | 1 368–1 436 | 307 | Proprietary |
| 56 | Grok 4.5 | xAI | 1 401 | 1 375–1 427 | 567 | Proprietary |
| 57 | GLM 4.6 | Z.ai | 1 401 | 1 375–1 426 | 599 | MIT |
| 58 | GPT-5.1 в режиме high | OpenAI | 1 399 | 1 376–1 423 | 651 | Proprietary |
| 59 | Claude Sonnet 4.5 | Anthropic | 1 399 | 1 382–1 416 | 1 302 | Proprietary |
| 60 | GLM 4.7 | Z.ai | 1 399 | 1 354–1 443 | 171 | MIT |
| 61 | MiMo-V2 Pro | Xiaomi | 1 399 | 1 370–1 427 | 456 | Proprietary |
| 62 | ChatGPT-4o (latest) | OpenAI | 1 398 | 1 382–1 415 | 1 355 | Proprietary |
| 63 | GPT-5.5 · 5.5-instant | OpenAI | 1 397 | 1 368–1 427 | 426 | Proprietary |
| 64 | Qwen3 Max · preview | Alibaba Qwen | 1 397 | 1 374–1 421 | 703 | Proprietary |
| 65 | Grok 4.1 · 4.1-thinking | xAI | 1 397 | 1 378–1 416 | 1 105 | Proprietary |
| 66 | o1 · 2024-12-17 | OpenAI | 1 396 | 1 364–1 428 | 396 | Proprietary |
| 67 | Qwen3 VL 235B A22B Instruct | Alibaba Qwen | 1 396 | 1 362–1 429 | 319 | Apache 2.0 |
| 68 | GLM 5V Turbo | Z.ai | 1 395 | 1 355–1 434 | 209 | Proprietary |
| 69 | GPT-5.1 | OpenAI | 1 393 | 1 371–1 416 | 719 | Proprietary |
| 70 | Amazon Nova Experimental Chat · 11-10 | Amazon | 1 393 | 1 363–1 423 | 388 | Proprietary |
| 71 | Qwen3 235B A22B Instruct 2507 | Alibaba Qwen | 1 393 | 1 378–1 408 | 1 786 | Apache 2.0 |
| 72 | Gemini 3.1 Flash Lite Preview | 1 392 | 1 373–1 412 | 1 049 | Proprietary | |
| 73 | GPT-4.5 Preview | OpenAI | 1 392 | 1 349–1 436 | 215 | Proprietary |
| 74 | GPT-5.2 Chat | OpenAI | 1 390 | 1 364–1 417 | 539 | Proprietary |
| 75 | Mistral Medium 3.5 | Mistral AI | 1 390 | 1 345–1 434 | 178 | Modified MIT |
| 76 | Gemini 2.5 Flash · flash-preview-09-2025 | 1 386 | 1 363–1 410 | 684 | Proprietary | |
| 77 | Gemini 2.5 Flash · flash | 1 386 | 1 373–1 400 | 2 344 | Proprietary | |
| 78 | Mistral Large 3 2512 | Mistral AI | 1 384 | 1 366–1 402 | 1 190 | Apache 2.0 |
| 79 | DeepSeek V4 Flash 0423 | DeepSeek | 1 383 | 1 362–1 405 | 869 | MIT |
| 80 | Qwen3.6 Plus | Alibaba Qwen | 1 383 | 1 362–1 405 | 791 | Proprietary |
| 81 | Nemotron 3 Ultra | NVIDIA | 1 383 | 1 339–1 427 | 202 | OpenMDW-1.1 |
| 82 | Qwen3.5 397B A17B | Alibaba Qwen | 1 382 | 1 365–1 400 | 1 314 | Apache 2.0 |
| 83 | Mistral Medium 3.1 | Mistral AI | 1 382 | 1 368–1 396 | 1 966 | Proprietary |
| 84 | GLM 4.5 | Z.ai | 1 381 | 1 357–1 406 | 639 | MIT |
| 85 | Claude Opus 4.1 · 20250805-thinking-16k | Anthropic | 1 381 | 1 362–1 400 | 1 095 | Proprietary |
| 86 | Qwen3 Max · 2025-09-23 | Alibaba Qwen | 1 378 | 1 340–1 415 | 253 | Proprietary |
| 87 | Kimi K2.5 · instant | Moonshot AI | 1 378 | 1 329–1 426 | 134 | Modified MIT |
| 88 | Grok 4 | xAI | 1 375 | 1 354–1 397 | 783 | Proprietary |
| 89 | Qwen3 235B A22B Thinking 2507 | Alibaba Qwen | 1 375 | 1 334–1 416 | 204 | Apache 2.0 |
| 90 | GPT-5.2 | OpenAI | 1 375 | 1 357–1 393 | 1 264 | Proprietary |
| 91 | Grok 3 | xAI | 1 374 | 1 347–1 401 | 520 | Proprietary |
| 92 | MiniMax M3 | MiniMax | 1 374 | 1 353–1 394 | 942 | MiniMax Community License |
| 93 | Claude Opus 4.5 в режиме high | Anthropic | 1 373 | 1 348–1 398 | 575 | Proprietary |
| 94 | ERNIE 5.0 · preview-1203 | Baidu | 1 372 | 1 328–1 417 | 159 | Proprietary |
| 95 | GPT-5.4 Mini в режиме high | OpenAI | 1 372 | 1 352–1 392 | 958 | Proprietary |
| 96 | LongCat-Flash-Chat · chat-2602-exp | Meituan | 1 372 | 1 344–1 399 | 470 | Proprietary |
| 97 | DeepSeek V3.2 · v3.2 | DeepSeek | 1 371 | 1 350–1 393 | 785 | MIT |
| 98 | o3 | OpenAI | 1 371 | 1 353–1 389 | 1 186 | Proprietary |
| 99 | DeepSeek V3.2 Exp · exp | DeepSeek | 1 370 | 1 331–1 410 | 211 | MIT |
| 100 | GPT-5.2 в режиме high | OpenAI | 1 370 | 1 347–1 392 | 773 | Proprietary |
| 101 | Qwen3 Next 80B A3B Instruct | Alibaba Qwen | 1 368 | 1 341–1 395 | 513 | Apache 2.0 |
| 102 | Claude Sonnet 4.5 в режиме high | Anthropic | 1 367 | 1 349–1 384 | 1 322 | Proprietary |
| 103 | MiMo-V2.5 | Xiaomi | 1 365 | 1 343–1 387 | 789 | MIT |
| 104 | Claude Opus 4.1 · 20250805 | Anthropic | 1 364 | 1 349–1 380 | 1 597 | Proprietary |
| 105 | Grok 4.1 Fast | xAI | 1 363 | 1 343–1 384 | 864 | Proprietary |
| 106 | GPT-5 | OpenAI | 1 361 | 1 340–1 383 | 806 | Proprietary |
| 107 | DeepSeek V3.2 · v3.2-thinking | DeepSeek | 1 360 | 1 337–1 384 | 687 | MIT |
| 108 | LongCat-Flash-Chat · chat | Meituan | 1 360 | 1 327–1 393 | 304 | MIT |
| 109 | Qwen3 235B A22B · a22b-no-thinking | Alibaba Qwen | 1 360 | 1 337–1 383 | 736 | Apache 2.0 |
| 110 | R1 0528 | DeepSeek | 1 359 | 1 324–1 393 | 321 | MIT |
| 111 | Grok 4 Fast · reasoning | xAI | 1 359 | 1 327–1 390 | 359 | Proprietary |
| 112 | Gemini 2.5 Flash Lite · 06-17-thinking | 1 358 | 1 336–1 381 | 725 | Proprietary | |
| 113 | MiMo-V2 Omni | Xiaomi | 1 358 | 1 323–1 393 | 294 | Proprietary |
| 114 | GPT-5 в режиме high | OpenAI | 1 358 | 1 336–1 379 | 833 | Proprietary |
| 115 | Grok 4.3 | xAI | 1 357 | 1 337–1 377 | 1 088 | Proprietary |
| 116 | Qwen3.5-27B | Alibaba Qwen | 1 357 | 1 328–1 386 | 443 | Apache 2.0 |
| 117 | MiMo-V2 Flash · flash (non-thinking) | Xiaomi | 1 357 | 1 335–1 379 | 759 | MIT |
| 118 | Inkling Small | Thinking Machines Lab | 1 356 | 1 325–1 386 | 376 | Apache 2.0 |
| 119 | DeepSeek V4 Flash 0423 в режиме high | DeepSeek | 1 355 | 1 334–1 376 | 913 | MIT |
| 120 | Qwen3.5-122B-A10B | Alibaba Qwen | 1 354 | 1 325–1 382 | 459 | Apache 2.0 |
| 121 | Qwen3.5-35B-A3B | Alibaba Qwen | 1 352 | 1 325–1 380 | 474 | Apache 2.0 |
| 122 | Qwen3 VL 235B A22B Thinking | Alibaba Qwen | 1 352 | 1 311–1 392 | 219 | Apache 2.0 |
| 123 | Hunyuan TurboS | Tencent | 1 351 | 1 301–1 402 | 133 | Proprietary |
| 124 | DeepSeek V3.2 Exp · exp-thinking | DeepSeek | 1 351 | 1 308–1 394 | 203 | MIT |
| 125 | GPT-5.3 Chat | OpenAI | 1 351 | 1 324–1 378 | 543 | Proprietary |
| 126 | Gemini 2.5 Flash Lite · 09-2025-no-thinking | 1 351 | 1 329–1 372 | 783 | Proprietary | |
| 127 | Step 3.5 Flash | StepFun | 1 348 | 1 329–1 367 | 1 048 | Apache 2.0 |
| 128 | Qwen3.5-Flash | Alibaba Qwen | 1 347 | 1 327–1 366 | 1 035 | Proprietary |
| 129 | Claude Haiku 4.5 | Anthropic | 1 343 | 1 328–1 358 | 2 066 | Proprietary |
| 130 | GPT-4.1 | OpenAI | 1 340 | 1 321–1 359 | 1 040 | Proprietary |
| 131 | DeepSeek V3.1 · v3.1 | DeepSeek | 1 337 | 1 310–1 364 | 490 | MIT |
| 132 | DeepSeek V3.1 · v3.1-thinking | DeepSeek | 1 336 | 1 305–1 368 | 373 | MIT |
| 133 | Grok 3 Mini в режиме high | xAI | 1 335 | 1 300–1 371 | 292 | Proprietary |
| 134 | Grok 4 Fast · chat | xAI | 1 334 | 1 291–1 376 | 187 | Proprietary |
| 135 | Qwen3 30B A3B Instruct 2507 | Alibaba Qwen | 1 333 | 1 308–1 358 | 584 | Apache 2.0 |
| 136 | Ling-flash-2.0 | inclusionAI | 1 332 | 1 294–1 371 | 231 | MIT |
| 137 | R1 | DeepSeek | 1 330 | 1 290–1 369 | 215 | MIT |
| 138 | Kimi K2 Thinking | Moonshot AI | 1 329 | 1 309–1 348 | 994 | Modified MIT |
| 139 | Amazon Nova Experimental Chat · 10-20 | Amazon | 1 328 | 1 283–1 373 | 179 | Proprietary |
| 140 | Gemini 2.0 Flash-Lite | 1 325 | 1 293–1 357 | 353 | Proprietary | |
| 141 | GPT-5.4 Nano в режиме high | OpenAI | 1 323 | 1 302–1 343 | 1 015 | Proprietary |
| 142 | Kimi K2 0905 | Moonshot AI | 1 322 | 1 287–1 357 | 314 | Modified MIT |
| 143 | DeepSeek V3 0324 | DeepSeek | 1 321 | 1 302–1 341 | 949 | MIT |
| 144 | Claude Opus 4 · 20250514-thinking-16k | Anthropic | 1 319 | 1 296–1 341 | 776 | Proprietary |
| 145 | Qwen3 235B A22B · a22b | Alibaba Qwen | 1 318 | 1 290–1 345 | 499 | Apache 2.0 |
| 146 | MiniMax M2.7 | MiniMax | 1 316 | 1 298–1 335 | 1 261 | Modified MIT |
| 147 | Gemini 2.0 Flash | 1 314 | 1 291–1 337 | 664 | Proprietary | |
| 148 | o3 Mini High | OpenAI | 1 314 | 1 273–1 354 | 241 | Proprietary |
| 149 | o4 Mini | OpenAI | 1 313 | 1 293–1 332 | 979 | Proprietary |
| 150 | Qwen3 Coder 480B A35B | Alibaba Qwen | 1 309 | 1 283–1 334 | 610 | Apache 2.0 |
| 151 | Qwen3 Next 80B A3B Thinking | Alibaba Qwen | 1 308 | 1 275–1 341 | 347 | Apache 2.0 |
| 152 | Gemma 3 27B | 1 307 | 1 287–1 328 | 881 | Gemma | |
| 153 | GLM 4.5 Air | Z.ai | 1 306 | 1 282–1 329 | 658 | MIT |
| 154 | GPT-5 Mini в режиме high | OpenAI | 1 304 | 1 280–1 329 | 654 | Proprietary |
| 155 | Grok 3 Mini | xAI | 1 304 | 1 274–1 334 | 392 | Proprietary |
| 156 | Qwen2.5 Max | Alibaba Qwen | 1 304 | 1 276–1 332 | 465 | Proprietary |
| 157 | Mistral Medium 3 | Mistral AI | 1 303 | 1 279–1 327 | 635 | Proprietary |
| 158 | Trinity Large Thinking | Arcee AI | 1 301 | 1 275–1 328 | 527 | Apache 2.0 |
| 159 | Claude Opus 4 · 20250514 | Anthropic | 1 301 | 1 281–1 320 | 1 014 | Proprietary |
| 160 | Gemini 1.5 Pro · 1.5-pro-002 | 1 298 | 1 274–1 321 | 773 | Proprietary | |
| 161 | GPT-4.1 Mini | OpenAI | 1 298 | 1 276–1 319 | 806 | Proprietary |
| 162 | Claude Sonnet 4 · 20250514 | Anthropic | 1 294 | 1 273–1 315 | 920 | Proprietary |
| 163 | MiniMax M2.1 | MiniMax | 1 292 | 1 256–1 329 | 279 | MIT |
| 164 | o1 · preview | OpenAI | 1 290 | 1 260–1 321 | 493 | Proprietary |
| 165 | Kimi K2 0711 | Moonshot AI | 1 287 | 1 263–1 312 | 648 | Modified MIT |
| 166 | MiMo-V2 Flash · flash (thinking) | Xiaomi | 1 286 | 1 243–1 330 | 161 | MIT |
| 167 | GLM 4.7 Flash | Z.ai | 1 285 | 1 244–1 326 | 235 | MIT |
| 168 | Command A | Cohere | 1 285 | 1 266–1 303 | 1 096 | CC-BY-NC-4.0 |
| 169 | Nova 2 Lite | Amazon | 1 284 | 1 234–1 333 | 156 | Proprietary |
| 170 | gpt-oss-120b | OpenAI | 1 282 | 1 258–1 306 | 668 | Apache 2.0 |
| 171 | QwQ 32B | Alibaba Qwen | 1 278 | 1 248–1 309 | 407 | Apache 2.0 |
| 172 | Claude Sonnet 4 · 20250514-thinking-32k | Anthropic | 1 272 | 1 249–1 295 | 723 | Proprietary |
| 173 | Step 3 | StepFun | 1 271 | 1 226–1 316 | 180 | Apache 2.0 |
| 174 | GPT-5 Nano в режиме high | OpenAI | 1 269 | 1 228–1 309 | 225 | Proprietary |
| 175 | Nemotron 3 Nano 30B A3B | NVIDIA | 1 269 | 1 228–1 309 | 236 | NVIDIA Open Model |
| 176 | Trinity Large | Arcee AI | 1 264 | 1 235–1 293 | 454 | Apache 2.0 |
| 177 | Qwen3 30B A3B | Alibaba Qwen | 1 262 | 1 235–1 289 | 472 | Apache 2.0 |
| 178 | MiniMax M1 | MiniMax | 1 262 | 1 237–1 286 | 659 | Apache 2.0 |
| 179 | Ring-flash-2.0 | inclusionAI | 1 261 | 1 219–1 302 | 218 | MIT |
| 180 | Mistral Small 3.2 24B | Mistral AI | 1 260 | 1 228–1 291 | 358 | Apache 2.0 |
| 181 | o3 Mini | OpenAI | 1 259 | 1 240–1 278 | 1 100 | Proprietary |
| 182 | Gemma 3n E4B | 1 259 | 1 232–1 285 | 512 | Gemma | |
| 183 | DeepSeek V3 | DeepSeek | 1 248 | 1 213–1 284 | 295 | DeepSeek |
| 184 | Claude 3.7 Sonnet · 20250219-thinking-32k | Anthropic | 1 248 | 1 225–1 272 | 760 | Proprietary |
| 185 | GLM-4-Plus | Z.ai | 1 243 | 1 211–1 276 | 374 | Proprietary |
| 186 | Claude 3.7 Sonnet · 20250219 | Anthropic | 1 239 | 1 216–1 263 | 700 | Proprietary |
| 187 | gpt-oss-20b | OpenAI | 1 239 | 1 198–1 280 | 226 | Apache 2.0 |
| 188 | Grok 2 | xAI | 1 237 | 1 215–1 259 | 952 | Proprietary |
| 189 | GPT-4o (2024-05-13) | OpenAI | 1 234 | 1 216–1 251 | 3 517 | Proprietary |
| 190 | MiniMax M2.5 | MiniMax | 1 232 | 1 208–1 257 | 709 | Modified MIT |
| 191 | Gemini 1.5 Pro · advanced-0514 | 1 232 | 1 211–1 253 | 1 945 | Proprietary | |
| 192 | Gemini 1.5 Pro · 1.5-pro-001 | 1 224 | 1 206–1 243 | 2 488 | Proprietary | |
| 193 | o1-mini | OpenAI | 1 223 | 1 199–1 247 | 716 | Proprietary |
| 194 | Gemini 1.5 Flash · 002 | 1 220 | 1 192–1 249 | 502 | Proprietary | |
| 195 | GPT-4o (2024-08-06) | OpenAI | 1 218 | 1 193–1 243 | 712 | Proprietary |
| 196 | DeepSeek V2.5 | DeepSeek | 1 209 | 1 176–1 243 | 344 | DeepSeek |
| 197 | OLMo 3.1 32B Instruct | Ai2 | 1 209 | 1 162–1 256 | 179 | Apache 2.0 |
| 198 | Llama 4 Scout | Meta | 1 208 | 1 183–1 233 | 652 | Llama |
| 199 | Athene V2 Chat | Nexusflow | 1 206 | 1 173–1 239 | 349 | NexusFlow |
| 200 | Mistral Small 3.1 24B | Mistral AI | 1 205 | 1 181–1 230 | 697 | Apache 2.0 |
| 201 | Llama 4 Maverick | Meta | 1 203 | 1 181–1 225 | 793 | Llama 4 |
| 202 | Nova Pro 1.0 | Amazon | 1 203 | 1 170–1 235 | 326 | Proprietary |
| 203 | Mistral Large · 2411 | Mistral AI | 1 202 | 1 172–1 232 | 408 | MRL |
| 204 | Gemini 1.5 Flash · 001 | 1 201 | 1 182–1 220 | 2 083 | Proprietary | |
| 205 | Claude 3.5 Sonnet · 20240620 | Anthropic | 1 200 | 1 181–1 219 | 1 715 | Proprietary |
| 206 | Claude 3.5 Sonnet · 20241022 | Anthropic | 1 197 | 1 178–1 216 | 1 270 | Proprietary |
| 207 | GPT-4o-mini (2024-07-18) | OpenAI | 1 195 | 1 175–1 215 | 1 087 | Proprietary |
| 208 | Grok 2 Mini | xAI | 1 195 | 1 171–1 218 | 852 | Proprietary |
| 209 | Yi-Lightning | 01.AI | 1 192 | 1 161–1 223 | 425 | Proprietary |
| 210 | Mistral Small 3 | Mistral AI | 1 188 | 1 149–1 228 | 231 | Apache 2.0 |
| 211 | Qwen2.5 72B Instruct | Alibaba Qwen | 1 188 | 1 160–1 216 | 527 | Qwen |
| 212 | Claude 3 Opus | Anthropic | 1 187 | 1 170–1 204 | 3 906 | Proprietary |
| 213 | GPT-4 Turbo | OpenAI | 1 187 | 1 168–1 205 | 2 668 | Proprietary |
| 214 | Llama 3.1 405B Instruct · fp8 | Meta | 1 184 | 1 162–1 206 | 924 | Llama 3.1 Community |
| 215 | GPT-4 Turbo Preview · 1106-preview | OpenAI | 1 184 | 1 163–1 205 | 1 292 | Proprietary |
| 216 | Gemma 2 27B | 1 174 | 1 155–1 193 | 1 440 | Gemma license | |
| 217 | GPT-4 Turbo Preview · 0125-preview | OpenAI | 1 173 | 1 152–1 194 | 1 393 | Proprietary |
| 218 | Llama 3.1 405B Instruct · bf16 | Meta | 1 172 | 1 146–1 198 | 586 | Llama 3.1 Community |
| 219 | Claude 3.5 Haiku | Anthropic | 1 172 | 1 154–1 191 | 1 206 | Proprietary |
| 220 | Mistral Large 2407 | Mistral AI | 1 170 | 1 146–1 194 | 751 | Mistral Research |
| 221 | Command R (08-2024) | Cohere | 1 162 | 1 111–1 213 | 152 | CC-BY-NC-4.0 |
| 222 | Aya Expanse 32B | Cohere | 1 158 | 1 127–1 189 | 385 | CC-BY-NC-4.0 |
| 223 | Nova Lite 1.0 | Amazon | 1 153 | 1 117–1 189 | 292 | Proprietary |
| 224 | Phi 4 | Microsoft | 1 151 | 1 118–1 184 | 319 | MIT |
| 225 | Athene 70B | Nexusflow | 1 150 | 1 119–1 182 | 396 | CC-BY-NC-4.0 |
| 226 | Nova Micro 1.0 | Amazon | 1 150 | 1 112–1 188 | 267 | Proprietary |
| 227 | Nemotron-4 340B Instruct | NVIDIA | 1 148 | 1 121–1 175 | 628 | NVIDIA Open Model |
| 228 | Llama 3.3 70B Instruct | Meta | 1 142 | 1 120–1 164 | 855 | Llama-3.3 |
| 229 | Qwen Max | Alibaba Qwen | 1 141 | 1 105–1 178 | 285 | Qwen |
| 230 | Llama 3.1 70B Instruct | Meta | 1 140 | 1 117–1 163 | 873 | Llama 3.1 Community |
| 231 | Gemini 1.5 Flash-8B | 1 140 | 1 111–1 169 | 500 | Proprietary | |
| 232 | Command R+ | Cohere | 1 138 | 1 118–1 158 | 1 919 | CC-BY-NC-4.0 |
| 233 | Gemma 2 9B | 1 137 | 1 116–1 158 | 1 087 | Gemma license | |
| 234 | Claude 3 Sonnet | Anthropic | 1 128 | 1 108–1 147 | 2 059 | Proprietary |
| 235 | Magistral Medium | Mistral AI | 1 125 | 1 077–1 173 | 194 | Proprietary |
| 236 | Claude 3 Haiku | Anthropic | 1 108 | 1 090–1 127 | 2 364 | Proprietary |
| 237 | Reka Flash 21B · 20240226-online | Reka AI | 1 104 | 1 069–1 139 | 309 | Proprietary |
| 238 | DeepSeek Coder V2 | DeepSeek | 1 104 | 1 072–1 136 | 399 | DeepSeek License |
| 239 | Command R | Cohere | 1 092 | 1 070–1 113 | 1 154 | CC-BY-NC-4.0 |
| 240 | GPT-4 · 0314 | OpenAI | 1 087 | 1 055–1 119 | 429 | Proprietary |
| 241 | Qwen2 72B Instruct | Alibaba Qwen | 1 083 | 1 061–1 104 | 1 167 | Qianwen LICENSE |
| 242 | Reka Flash 21B · 20240226 | Reka AI | 1 068 | 1 041–1 095 | 629 | Proprietary |
| 243 | GPT-4 · 0613 | OpenAI | 1 060 | 1 037–1 082 | 1 154 | Proprietary |
| 244 | Mixtral 8x22B Instruct | Mistral AI | 1 056 | 1 035–1 078 | 1 187 | Apache 2.0 |
| 245 | Gemma 2 2B | 1 055 | 1 029–1 080 | 717 | Gemma license | |
| 246 | Llama 3.1 8B Instruct | Meta | 1 053 | 1 029–1 077 | 802 | Llama 3.1 Community |
| 247 | Qwen1.5 72B Chat | Alibaba Qwen | 1 049 | 1 021–1 077 | 552 | Qianwen LICENSE |
| 248 | Qwen1.5 110B Chat | Alibaba Qwen | 1 043 | 1 022–1 065 | 1 434 | Qianwen LICENSE |
| 249 | GLM-4 | Z.ai | 1 042 | 1 006–1 078 | 302 | Proprietary |
| 250 | Mistral Medium (2023) | Mistral AI | 1 032 | 996–1 068 | 268 | Proprietary |
| 251 | GPT-3.5 Turbo | OpenAI | 1 019 | 996–1 041 | 1 184 | Proprietary |
| 252 | Mistral Large · 2402 | Mistral AI | 1 017 | 994–1 040 | 1 090 | Proprietary |
| 253 | Llama 3 70B Instruct | Meta | 1 017 | 999–1 034 | 5 410 | Llama 3 Community |
| 254 | Qwen1.5 32B Chat | Alibaba Qwen | 1 007 | 976–1 039 | 395 | Qianwen LICENSE |
| 255 | Yi-1.5 34B Chat | 01.AI | 1 005 | 980–1 030 | 804 | Apache-2.0 |
| 256 | Llama 3 8B Instruct | Meta | 1 003 | 984–1 022 | 2 422 | Llama 3 Community |
| 257 | DBRX Instruct | Databricks | 993 | 962–1 024 | 442 | DBRX LICENSE |
| 258 | Gemma 1.1 7B | 988 | 963–1 013 | 770 | Gemma license | |
| 259 | Mixtral 8x7B Instruct | Mistral AI | 968 | 942–993 | 778 | Apache 2.0 |
| 260 | Llama 2 70B Chat | Meta | 964 | 924–1 003 | 243 | Llama 2 Community |
| 261 | Yi 34B Chat | 01.AI | 959 | 918–1 000 | 227 | Yi License |
| 262 | Phi-3 Medium 4K Instruct | Microsoft | 954 | 929–979 | 732 | MIT |
| 263 | Llama 2 13B Chat | Meta | 953 | 903–1 004 | 172 | Llama 2 Community |
| 264 | Snowflake Arctic Instruct | Snowflake | 939 | 910–968 | 547 | Apache 2.0 |
| 265 | Phi-3 Mini 4K Instruct · instruct | Microsoft | 905 | 880–930 | 860 | MIT |
| 266 | Gemma 1.1 2B | 899 | 866–932 | 386 | Gemma license | |
| 267 | Phi-3 Small 8K Instruct | Microsoft | 894 | 868–919 | 726 | MIT |
| 268 | Phi-3 Mini 4K Instruct · instruct-june-2024 | Microsoft | 880 | 839–920 | 292 | MIT |
| 269 | Phi-3 Mini 128K Instruct | Microsoft | 856 | 816–896 | 260 | MIT |
Данные на 13 сентября 2026. Как мы считаем: методика
Свежесть данных
Снимок замеров опубликован 13 сентября 2026; всего снимков борда в архиве: 200.
- замеры оценок людей: загружено 18 сентября 2026, 08:25 UTC
- каталог моделей и цены: загружено 18 сентября 2026, 08:51 UTC
- доли использования: загружено 18 сентября 2026, 08:51 UTC
Другие рейтинги
- Рейтинг агентов
- Рейтинг работы с документами
- Рейтинг редактирования изображений
- Рейтинг видео из изображения
- Популярность моделей по использованию через API-агрегаторы
- Рейтинг поисковых моделей
- Рейтинг моделей для кода
- Рейтинг моделей для творческих текстов
- Рейтинг моделей для математики
- Рейтинг моделей для русского языка
- Рейтинг генерации изображений
- Рейтинг генерации видео
- Рейтинг редактирования видео
- Рейтинг моделей с изображениями на входе
- Рейтинг моделей для веб-разработки
Данные и источники
Обновлено 18 сентября 2026. Каталог, цены и режимы проверяются ежедневно по нескольким API-агрегаторам и провайдерам. Места — по открытым замерам оценок людей, интервалы приведены как в замере.