Qwen Models

149 modelsGeneral models from $0.0206/M inputUp to 1.05M context

Usage

Last 30 days · 2026-09-03 to 2026-10-02

Tokens

256B

Requests

6.9M

Models in use

140 of 149

Tokens per day, stacked by model

010.5B21.1B09-0309-1009-1709-2410-012026-09-03 — 5,537,612,300 tokens qwen3.8-flash: 2,739,642,180 qwen3.7-flash: 998,547,835 qwen3.8-max-2026-09-02: 906,928,385 90 more models: 527,356,220 qwen3.8-max: 279,175,850 qwen3.5-35b-a3b: 43,131,965 qwen3.7-plus: 42,829,8652026-09-04 — 12,488,881,260 tokens qwen3.8-flash: 9,993,403,840 90 more models: 1,070,554,880 qwen3.7-flash: 524,029,925 qwen3.8-max: 412,862,265 qwen3.8-max-2026-09-02: 368,870,925 qwen3.7-plus: 76,676,335 qwen3.5-35b-a3b: 42,483,000 qwen3-coder-30b-a3b-instruct: 902026-09-05 — 4,022,671,485 tokens qwen3.8-flash: 2,243,267,610 qwen3.8-max: 625,610,780 qwen3.7-flash: 456,534,395 90 more models: 419,082,515 qwen3.8-max-2026-09-02: 121,533,475 qwen3.7-plus: 113,934,270 qwen3.5-35b-a3b: 42,708,250 qwen3-coder-30b-a3b-instruct: 1902026-09-06 — 2,626,546,915 tokens qwen3.8-flash: 783,405,980 90 more models: 552,940,645 qwen3.7-flash: 525,941,750 qwen3.8-max: 389,019,335 qwen3.8-max-2026-09-02: 213,127,295 qwen3.7-plus: 108,501,455 qwen3.5-35b-a3b: 53,610,335 qwen3-coder-30b-a3b-instruct: 1202026-09-07 — 9,239,671,715 tokens qwen3.8-flash: 6,190,494,710 qwen3.7-flash: 1,257,714,380 90 more models: 759,452,445 qwen3.8-max: 454,783,380 qwen3.7-plus: 332,139,620 qwen3.8-max-2026-09-02: 206,455,505 qwen3.5-35b-a3b: 38,631,505 qwen3-coder-30b-a3b-instruct: 1702026-09-08 — 5,829,326,175 tokens qwen3.7-flash: 2,068,261,435 qwen3.8-flash: 1,577,516,590 90 more models: 1,124,568,070 qwen3.8-max: 622,859,425 qwen3.7-plus: 328,252,005 qwen3.8-max-2026-09-02: 67,832,015 qwen3.5-35b-a3b: 39,886,465 qwen3-coder-30b-a3b-instruct: 150,1702026-09-09 — 5,185,373,400 tokens qwen3.8-flash: 1,970,807,585 90 more models: 1,386,167,950 qwen3.7-flash: 799,413,475 qwen3.8-max: 423,545,040 qwen3.7-plus: 351,726,740 qwen3.8-max-2026-09-02: 206,650,860 qwen3.5-35b-a3b: 47,043,485 qwen3-coder-30b-a3b-instruct: 18,2652026-09-10 — 5,191,505,810 tokens qwen3.7-flash: 1,975,783,065 90 more models: 1,001,289,925 qwen3.7-plus: 630,277,955 qwen3.8-flash: 626,499,440 qwen3.8-max-2026-09-02: 473,250,630 qwen3.8-max: 446,450,955 qwen3.5-35b-a3b: 37,953,590 qwen3-coder-30b-a3b-instruct: 2502026-09-11 — 8,244,349,505 tokens qwen3.7-flash: 4,781,081,620 90 more models: 1,536,143,245 qwen3.8-flash: 550,154,925 qwen3.8-max: 538,164,185 qwen3.8-max-2026-09-02: 325,398,580 qwen3.7-plus: 298,764,370 qwen3.5-35b-a3b: 214,642,5802026-09-12 — 3,545,208,675 tokens 90 more models: 1,021,319,590 qwen3.8-max: 826,539,465 qwen3.7-flash: 766,822,830 qwen3.8-max-2026-09-02: 423,241,700 qwen3.8-flash: 270,429,405 qwen3.7-plus: 186,075,810 qwen3.5-35b-a3b: 50,779,705 qwen3-coder-30b-a3b-instruct: 1702026-09-13 — 5,054,764,075 tokens qwen3.8-max: 1,247,804,290 qwen3.7-flash: 1,210,605,070 90 more models: 894,981,935 qwen3.8-flash: 821,153,755 qwen3.8-max-2026-09-02: 609,629,460 qwen3.7-plus: 243,396,330 qwen3.5-35b-a3b: 27,193,145 qwen3-coder-30b-a3b-instruct: 902026-09-14 — 7,398,658,815 tokens qwen3.8-flash: 2,362,250,595 qwen3.7-flash: 1,747,887,370 90 more models: 1,428,794,890 qwen3.8-max: 1,127,928,580 qwen3.8-max-2026-09-02: 607,695,115 qwen3.5-35b-a3b: 100,327,290 qwen3.7-plus: 23,774,135 qwen3-coder-30b-a3b-instruct: 8402026-09-15 — 8,960,980,510 tokens qwen3.8-max: 4,256,658,605 qwen3.7-flash: 1,843,092,255 90 more models: 1,632,657,900 qwen3.8-flash: 684,446,835 qwen3.5-35b-a3b: 294,812,400 qwen3.8-max-2026-09-02: 163,879,425 qwen3.7-plus: 85,433,0902026-09-16 — 7,173,950,185 tokens 90 more models: 2,285,622,500 qwen3.7-flash: 1,763,559,090 qwen3.8-max: 1,221,507,235 qwen3.8-flash: 996,883,320 qwen3.5-35b-a3b: 834,655,585 qwen3.8-max-2026-09-02: 57,504,095 qwen3.7-plus: 14,217,910 qwen3-coder-30b-a3b-instruct: 4502026-09-17 — 8,577,738,830 tokens qwen3.8-flash: 3,696,625,615 qwen3.7-flash: 2,619,348,795 90 more models: 977,417,880 qwen3.8-max: 638,809,955 qwen3.7-plus: 475,708,395 qwen3.5-35b-a3b: 142,311,500 qwen3.8-max-2026-09-02: 27,516,610 qwen3-coder-30b-a3b-instruct: 802026-09-18 — 17,347,481,805 tokens qwen3.8-flash: 12,710,425,175 qwen3.8-max: 1,610,283,420 qwen3.7-flash: 1,132,189,755 90 more models: 879,506,480 qwen3.7-plus: 559,943,425 qwen3.8-max-2026-09-02: 305,132,615 qwen3.8-omni-flash: 103,537,690 qwen3.5-35b-a3b: 46,458,720 qwen3-coder-30b-a3b-instruct: 4,5252026-09-19 — 8,982,063,990 tokens qwen3.8-flash: 4,082,188,335 qwen3.7-flash: 1,665,555,715 qwen3.8-omni-flash: 905,188,720 qwen3.7-plus: 611,623,125 90 more models: 591,984,235 qwen3.8-max-2026-09-02: 478,414,715 qwen3.8-max: 440,095,410 qwen3.5-35b-a3b: 207,011,505 qwen3-coder-30b-a3b-instruct: 2,2302026-09-20 — 11,191,614,860 tokens qwen3.8-flash: 4,649,415,195 qwen3.7-flash: 1,861,462,445 qwen3.5-35b-a3b: 1,505,951,595 qwen3.8-max-2026-09-02: 1,149,355,110 qwen3.8-max: 867,399,815 qwen3.7-plus: 675,817,580 90 more models: 345,764,215 qwen3.8-omni-flash: 136,448,9052026-09-21 — 14,041,612,085 tokens qwen3.8-max: 7,060,875,885 qwen3.8-flash: 2,386,480,075 90 more models: 1,156,142,475 qwen3.5-35b-a3b: 1,127,355,685 qwen3.8-omni-flash: 818,181,770 qwen3.7-flash: 665,805,600 qwen3.7-plus: 576,448,760 qwen3.8-max-2026-09-02: 250,321,8352026-09-22 — 11,646,872,895 tokens qwen3.8-flash: 4,606,537,145 qwen3.5-35b-a3b: 2,074,607,095 qwen3.8-max: 2,010,537,490 qwen3.7-flash: 1,346,858,120 qwen3.7-plus: 721,761,690 qwen3.8-omni-flash: 407,123,735 90 more models: 382,029,530 qwen3.8-max-2026-09-02: 97,415,690 qwen3-coder-30b-a3b-instruct: 2,4002026-09-23 — 12,287,628,245 tokens qwen3.8-flash: 5,739,027,535 qwen3.8-max: 2,385,397,500 90 more models: 1,185,687,490 qwen3.7-flash: 844,419,445 qwen3.5-35b-a3b: 791,707,345 qwen3.8-max-2026-09-02: 503,298,145 qwen3-coder-30b-a3b-instruct: 490,868,810 qwen3.7-plus: 281,089,645 qwen3.8-omni-flash: 66,132,3302026-09-24 — 20,195,964,720 tokens qwen3.8-flash: 11,504,040,330 qwen3-coder-30b-a3b-instruct: 3,006,674,535 qwen3.8-max: 2,745,371,645 90 more models: 2,395,643,935 qwen3.8-max-2026-09-02: 301,598,275 qwen3.7-flash: 125,287,835 qwen3.8-omni-flash: 46,731,250 qwen3.5-35b-a3b: 44,402,025 qwen3.7-plus: 26,214,8902026-09-25 — 21,082,856,275 tokens qwen3.8-max: 9,398,514,180 qwen3.8-flash: 8,188,988,460 qwen3-coder-30b-a3b-instruct: 2,027,577,270 90 more models: 666,828,710 qwen3.7-flash: 401,386,750 qwen3.8-max-2026-09-02: 244,008,240 qwen3.5-35b-a3b: 62,913,890 qwen3.8-omni-flash: 51,684,600 qwen3.7-plus: 40,954,1752026-09-26 — 3,981,226,495 tokens qwen3-coder-30b-a3b-instruct: 1,696,651,555 qwen3.8-flash: 1,338,431,375 90 more models: 361,897,525 qwen3.7-flash: 230,181,860 qwen3.5-35b-a3b: 196,820,990 qwen3.8-max-2026-09-02: 54,303,015 qwen3.8-omni-flash: 53,582,080 qwen3.8-max: 33,601,750 qwen3.7-plus: 15,756,3452026-09-27 — 2,151,931,340 tokens qwen3.8-flash: 1,170,150,275 qwen3.7-flash: 328,442,035 qwen3.8-max-2026-09-02: 314,734,910 90 more models: 161,071,895 qwen3.8-omni-flash: 64,152,605 qwen3.5-35b-a3b: 55,108,095 qwen3.8-max: 52,716,635 qwen3.7-plus: 5,554,8902026-09-28 — 10,634,926,650 tokens 90 more models: 5,139,118,710 qwen3.8-flash: 4,070,990,445 qwen3.7-flash: 412,477,350 qwen3.8-max: 352,693,960 qwen3.8-omni-flash: 345,521,465 qwen3.7-plus: 234,994,725 qwen3.5-35b-a3b: 47,184,945 qwen3.8-max-2026-09-02: 31,944,635 qwen3-coder-30b-a3b-instruct: 4152026-09-29 — 4,546,403,745 tokens qwen3.8-flash: 1,400,724,605 90 more models: 1,323,903,595 qwen3.5-35b-a3b: 585,889,460 qwen3.8-omni-flash: 560,086,750 qwen3.7-flash: 460,119,670 qwen3.8-max: 129,189,875 qwen3.7-plus: 62,115,215 qwen3.8-max-2026-09-02: 24,374,460 qwen3-coder-30b-a3b-instruct: 1152026-09-30 — 8,256,254,320 tokens qwen3.8-flash: 4,085,519,925 qwen3.8-omni-flash: 1,886,860,200 90 more models: 1,322,317,570 qwen3.7-flash: 374,012,810 qwen3.8-max: 292,055,005 qwen3.5-35b-a3b: 184,877,665 qwen3.7-plus: 71,778,085 qwen3.8-max-2026-09-02: 38,833,0602026-10-01 — 4,469,398,660 tokens qwen3.8-flash: 1,609,646,190 qwen3.8-omni-flash: 1,411,684,980 qwen3.8-max: 555,115,995 qwen3.7-flash: 495,695,145 90 more models: 358,209,030 qwen3.5-35b-a3b: 28,189,815 qwen3.7-plus: 10,675,910 qwen3.8-max-2026-09-02: 181,5952026-10-02 — 5,860,575,690 tokens qwen3.8-flash: 3,905,171,175 90 more models: 591,904,020 qwen3.8-max: 550,423,985 qwen3.7-flash: 425,506,845 qwen3.8-omni-flash: 228,133,805 qwen3.8-max-2026-09-02: 91,669,880 qwen3.7-plus: 45,007,665 qwen3.5-35b-a3b: 22,758,315
  • qwen3.8-flash
  • qwen3.8-max
  • qwen3.7-flash
  • qwen3.5-35b-a3b
  • qwen3.8-max-2026-09-02
  • qwen3.7-plus
  • qwen3-coder-30b-a3b-instruct
  • qwen3.8-omni-flash
  • 90 more models

Which models that traffic went to

  1. Qwen3.8 Flash41.8%107B
  2. Qwen3.8 Max16.4%42B
  3. Qwen3.7 Flash13.3%34.1B
  4. Qwen3.5 35B A3B3.5%9B
  5. Qwen3.8 Max 2026 09-023.4%8.7B
  6. Qwen3.7 Plus2.8%7.3B
  7. Qwen3 Coder 30B A3B Instruct2.8%7.2B
  8. Qwen3.8 Omni Flash2.8%7.1B
  9. 90 more models13.1%33.5B

Share of 256B tokens. 42 models with traffic report no token counts and cannot be ranked here, including qwen-audio-3.0-tts-flash and qwen-audio-3.0-tts-plus — they are in the request view.

The two views disagree on purpose: a model can take a large share of the calls and a small share of the tokens — many short requests — or the reverse. Which one matters depends on whether your cost is driven by call volume or by prompt length. Measured on AIHubMix over the last 30 days, counting the 149 model IDs listed on this page; traffic routed through upstream-specific IDs that are not in the public catalog is not included.

All 149 Qwen Models

Open in model list
Qwen models on AIHubMix with input and output modalities, context length, maximum output, price per million tokens including cache read and cache write rates, and measured throughput and latency.
Modalities
qwen3-coder-plusTakes text, returns text.1.05M66K$0.54$2.16/M$0.108/M—83 tok/s1.63 s
qwen3.6-plus-preview-freeTakes text, returns text.1M66KFreeFree/M————
qwen3.5-flashTakes text, vision, video, returns text.1M66K$0.0282$0.282/M$0.0028/M$0.0352/M62 tok/s1.71 s
qwen3.7-flashTakes text, vision, video, returns text.1M131K$0.0282$0.1128/M$0.0056/M$0.0352/M76 tok/s7.06 s
qwen3.5-plusTakes text, vision, video, returns text.1M66K$0.1096$0.6576/M$0.011/M$0.137/M49 tok/s1.01 s
qwen3.8-flashTakes text, vision, video, returns text.1M131K$0.1126$0.38/M$0.0141/M$0.1759/M39 tok/s23.17 s
qwen3.8-omni-flashTakes text, vision, audio, video, returns text.1M131K$0.1126$0.38/M$0.0141/M$0.1759/M18 tok/s24.57 s
qwen3-coder-flashTakes text, returns text.1M66K$0.136$0.544/M——110 tok/s1.85 s
qwen3.6-flashTakes text, vision, video, returns text.1M66K$0.169$1.014/M$0.0169/M$0.2112/M30 tok/s22.67 s
qwen3.8-27bTakes text, vision, video, returns text.1M131K$0.2$2.5/M$0.05/M———
qwen3.6-plusTakes text, vision, video, returns text.1M66K$0.282$1.692/M$0.0282/M$0.3525/M36 tok/s6.79 s
qwen3.7-plusTakes text, vision, video, returns text.1M131K$0.282$1.128/M$0.0564/M$0.3525/M34 tok/s4.05 s
qwen3.8-max-previewTakes text, vision, video, returns text.1M131K$0.338$1.014/M$0.0676/M$0.4225/M48 tok/s2.40 s
qwen3.7-maxTakes text, returns text.1M131K$1.69$5.07/M$0.169/M$2.1125/M69 tok/s3.27 s
qwen3.8-maxTakes text, vision, video, returns text.1M131K$1.69$5.07/M$0.169/M$2.1125/M37 tok/s13.79 s
qwen3.8-max-2026-09-02Takes text, vision, video, returns text.1M131K$1.69$5.07/M$0.169/M$2.1125/M24 tok/s12.10 s
qwen3.8-2.4t-a95bTakes text, returns text.1M131K$2$6/M$0.5/M—48 tok/s2.40 s
qwen3-vl-flashTakes text, vision, video, returns text.262K33K$0.0206$0.206/M$0.0041/M—49 tok/s1.96 s
qwen3-vl-flash-2026-01-22Takes text, vision, video, returns text.262K33K$0.0206$0.206/M——24 tok/s0.33 s
qwen3.5-35b-a3bTakes text, vision, video, returns text.262K66K$0.0564$0.4512/M——71 tok/s1.27 s
qwen3.5-27bTakes text, vision, video, returns text.262K66K$0.0846$0.6768/M——26 tok/s2.69 s
qwen3.5-122b-a10bTakes text, vision, video, returns text.262K66K$0.1126$0.9008/M——71 tok/s1.27 s
qwen3-coder-nextTakes text, returns text.262K66K$0.137$0.548/M——16 tok/s0.51 s
qwen3-vl-plusTakes text, vision, video, returns text.262K33K$0.137$1.37/M$0.0274/M—67 tok/s2.32 s
qwen3.5-397b-a17bTakes text, vision, video, returns text.262K66K$0.1644$0.9864/M——46 tok/s1.96 s
qwen3-coder-30b-a3b-instructTakes text, returns text.262K262K$0.2$0.8/M——115 tok/s0.95 s
qwen3.6-35b-a3bTakes text, vision, video, returns text.262K66K$0.254$1.524/M——33 tok/s2.62 s
qwen3-235b-a22b-instruct-2507Takes text, vision, returns text.262K16K$0.28$1.12/M——96 tok/s1.03 s
qwen3-235b-a22b-thinking-2507Takes text, vision, returns text.262K131K$0.28$2.8/M——87 tok/s0.27 s
qwen3-235b-a22b-2507Takes text, returns text.262K—$0.35$1.4/M$0.07/M———
qwen3.6-27bTakes text, vision, video, returns text.262K66K$0.422$2.532/M——81 tok/s1.84 s
qwen3-maxTakes text, returns text.262K66K$0.4508$1.8032/M$0.0902/M$0.5635/M35 tok/s1.21 s
qwen3-max-2026-01-23Takes text, returns text.262K66K$0.4508$1.8032/M$0.0902/M$0.5635/M——
qwen3.6-max-previewTakes text, returns text.262K66K$1.268$7.608/M$0.1268/M$1.585/M185 tok/s0.65 s
qwen3-coder-480b-a35b-instructTakes text, returns text.262K66K$0.82$3.28/M——1655 tok/s0.92 s
qwen3-next-80b-a3b-instructTakes text, vision, returns text.256K33K$0.138$0.552/M——150 tok/s0.10 s
qwen3-next-80b-a3b-thinkingTakes text, vision, returns text.256K33K$0.142$1.42/M——221 tok/s0.50 s
qwen3-235b-a22bTakes text, returns text.131K128K$0.28$1.12/M——81 tok/s0.63 s
qwen3-vl-30b-a3b-instructTakes text, vision, video, returns text.131K33K$0.1028$0.4112/M——42 tok/s1.07 s
qwen3-vl-30b-a3b-thinkingTakes text, vision, video, returns text.131K33K$0.1028$1.028/M——42 tok/s1.49 s
qwen3-vl-235b-a22b-instructTakes text, vision, video, returns text.131K33K$0.274$1.096/M——57 tok/s1.13 s
qwen3-vl-235b-a22b-thinkingTakes text, vision, video, returns text.131K33K$0.274$2.74/M——58 tok/s2.22 s
qwen-3.8-27bTakes text, vision, video, returns text.131K—$1.1$1.65/M——710 tok/s0.59 s
bai-qwen3-vl-235b-a22b-instructTakes , returns text.131K—$0.274$1.096/M——13 tok/s0.50 s
kat-devTakes text, returns text.128K—$0.137$0.548/M————
Qwen/Qwen2.5-VL-72B-InstructTakes text, vision, video, returns text.128K—$0.5$0.5/M————
qwen3-coder-plus-2025-07-22Takes text, returns text.128K66K$0.54$2.16/M$0.108/M—83 tok/s1.63 s
decision-model-previewTakes text. Output modality not published.64K—FreeFree/M————
qwen3-reranker-0.6bTakes text, vision. Output modality not published.16K8K$0.11$0.11/M————
qwen-mt-turboTakes text, returns text.16K8K$0.192$0.5349/M——6 tok/s1.12 s
qwen-mt-plusTakes text, returns text.16K8K$0.492$1.476/M——7 tok/s1.00 s
qwen-audio-3.0-tts-flashTakes text, returns audio.——Free$0.141/M————
qwen-audio-3.0-tts-plusTakes text, returns audio.——Free$0.197/M————
qwen-image-3.0Takes text, vision, returns vision.——FreeFree/M————
qwen-image-3.0-proTakes text, vision, returns vision.——FreeFree/M————
qwen-flash——$0.02$0.2/M——4 tok/s2.00 s
qwen-flash-2025-07-28——$0.02$0.2/M————
qwen-turboTakes text, returns text.——$0.046$0.092/M$0.0092/M—8 tok/s0.70 s
qwen-turbo-2024-11-01Takes text, returns text.——$0.046$0.092/M——94 tok/s0.66 s
qwen-turbo-2025-04-28Takes , returns text.——$0.046$0.092/M————
qwen-turbo-latestTakes , returns text.——$0.046$0.092/M$0.0092/M———
qwen3-0.6bTakes , returns text.——$0.046$0.46/M————
qwen3-1.7bTakes , returns text.——$0.046$0.46/M————
qwen3-4bTakes , returns text.——$0.046$0.46/M————
bce-reranker-baseTakes text, vision. Output modality not published.——$0.068$0.068/M————
qwen3-embedding-0.6bTakes text. Output modality not published.——$0.068$0.068/M————
qwen3-embedding-4bTakes text. Output modality not published.——$0.068$0.068/M————
qwen3-embedding-8bTakes text. Output modality not published.——$0.068$0.068/M————
Qwen/Qwen2-7B-Instruct——$0.08$0.08/M————
qwen3-8bTakes , returns text.——$0.08$0.8/M——64 tok/s1.11 s
text-embedding-v4Takes text. Output modality not published.——$0.08$0.08/M————
qwen-long——$0.1$0.4/M——40 tok/s1.08 s
qwen3-30b-a3b-instruct-2507——$0.1028$0.4112/M————
gte-rerank-v2Takes text, vision. Output modality not published.——$0.11$0.11/M————
qwen3-reranker-4bTakes text, vision. Output modality not published.——$0.11$0.11/M————
qwen3-reranker-8bTakes text, vision. Output modality not published.——$0.11$0.11/M————
qwen-plus——$0.1126$1.126/M$0.0225/M$0.1407/M59 tok/s0.74 s
qwen-plus-2025-04-28Takes , returns text.——$0.1126$1.126/M$0.0225/M$0.1407/M5 tok/s1.28 s
qwen-plus-2025-07-28——$0.1126$1.126/M$0.0225/M$0.1407/M——
qwen-plus-latestTakes , returns text.——$0.1126$1.126/M$0.0225/M$0.1407/M9 tok/s0.67 s
qwen3-30b-a3b——$0.12$1.2/M————
qwen3-30b-a3b-thinking-2507——$0.12$1.2/M————
gme-qwen2-vl-2b-instructTakes text, vision, video. Output modality not published.——$0.138$0.138/M————
Qwen/QwQ-32BTakes , returns text.——$0.14$0.56/M————
Qwen/Qwen2.5-Coder-32B-Instruct——$0.16$0.16/M————
Qwen/QwQ-32B-Preview——$0.16$0.16/M————
qwen3-14bTakes , returns text.——$0.16$1.6/M——47 tok/s0.34 s
qwen3-32b——$0.16$0.64/M————
Qwen/Qwen2-1.5B-Instruct——$0.2$0.2/M————
Qwen/Qwen3-8B——$0.2$0.2/M——11 tok/s3.31 s
qwen2.5-coder-1.5b-instruct——$0.2$0.4/M————
qwen2.5-coder-7b-instruct——$0.2$0.4/M————
qwen2.5-math-1.5b-instruct——$0.2$0.2/M————
qwen2.5-math-7b-instruct——$0.2$0.4/M————
Qwen/Qwen2-57B-A14B-Instruct——$0.24$0.24/M————
Qwen/Qwen2.5-VL-32B-InstructTakes text, vision, video, returns text.——$0.24$0.24/M————
qwen-3-235b-a22b-instruct-2507——$0.28$1.4/M————
qwen-3-235b-a22b-thinking-2507——$0.28$2.8/M————
Qwen2-VL-7B-InstructTakes text, vision, video. Output modality not published.——$0.28$0.7/M————
Qwen3-235B-A22B-Thinking-2507——$0.28$2.8/M——87 tok/s0.27 s
qwen-max——$0.38$1.52/M——28 tok/s0.41 s
qwen-max-0125——$0.38$1.52/M————
qwen-3-32b——$0.4$1.6/M————
qwen-qwq-32b——$0.4$0.8/M————
Qwen/Qwen2.5-7B-Instruct——$0.4$0.4/M————
Qwen/Qwen3-32B——$0.4$0.8/M————
qwen2.5-14b-instruct——$0.4$1.2/M————
qwen2.5-3b-instruct——$0.4$0.8/M————
qwen2.5-7b-instruct——$0.4$0.8/M————
Qwen/Qwen3-14B——$0.5$0.5/M————
Qwen/Qwen2.5-32B-Instruct——$0.6$0.6/M——24 tok/s1.62 s
qwen2.5-32b-instruct——$0.6$1.2/M————
Qwen/Qwen2-72B-Instruct——$0.8$0.8/M————
Qwen/Qwen2.5-72B-Instruct——$0.8$0.8/M————
Qwen/Qwen2.5-72B-Instruct-128K——$0.8$0.8/M————
qwen2.5-72b-instruct——$0.8$2.4/M————
qwen2.5-math-72b-instruct——$0.8$2.4/M————
qwen3-max-previewTakes text, vision, returns text.——$0.846$3.384/M$0.1692/M—8 tok/s0.63 s
Qwen/Qwen3-30B-A3B——$1$1/M————
Qwen/QVQ-72B-Preview——$1.2$1.2/M————
qwen-imageTakes text, vision, returns vision.——$2$2/M————
qwen-image-2.0Takes text, vision, returns vision.——$2$2/M————
qwen-image-2.0-proTakes text, vision, returns vision.——$2$2/M————
qwen-image-editTakes text, vision, returns vision.——$2$2/M————
qwen-image-maxTakes text, vision, returns vision.——$2$2/M————
wan2.7-imageTakes text, vision, returns vision.——$2$2/M————
wan2.7-image-proTakes text, vision, returns vision.——$2$2/M————
Qwen2-VL-72B-InstructTakes text, vision, video. Output modality not published.——$2.18$6.54/M————
qwen2.5-vl-72b-instructTakes text, vision, returns text.——$2.4$7.2/M————
qwen-max-longcontext——$7$21/M————
wan2.2-i2v-plusTakes text, vision, returns video.——$480$480/M————
wan2.5-i2v-previewTakes text, vision, returns video.——$480$480/M————
wan2.5-t2v-previewTakes text, returns video.——$480$480/M————
wan3.0-videoTakes text, vision, audio, video, returns video.——$480$480/M————
wan3.0-video-primeTakes text, vision, audio, video, returns video.——$480$480/M————
happyhorse-1.0-i2vTakes text, vision, returns video.——$720$720/M————
happyhorse-1.0-r2vTakes text, vision, returns video.——$720$720/M————
happyhorse-1.0-t2vTakes text, returns video.——$720$720/M————
happyhorse-1.0-video-editTakes text, vision, video, returns video.——$720$720/M————
happyhorse-1.1-i2vTakes text, vision, returns video.——$720$720/M————
happyhorse-1.1-r2vTakes text, vision, returns video.——$720$720/M————
happyhorse-1.1-t2vTakes text, returns video.——$720$720/M————
wan2.6-i2vTakes text, vision, returns video.——$720$720/M————
wan2.6-t2iTakes text, vision, returns vision.——$720$720/M————
wan2.6-t2vTakes text, returns video.——$720$720/M————
wan2.7-i2vTakes text, vision, audio, returns video.——$720$720/M————
wan2.7-r2vTakes text, vision, audio, video, returns video.——$720$720/M————
wan2.7-t2vTakes text, audio, returns video.——$720$720/M————
wan2.7-videoeditTakes text, vision, video, returns video.——$720$720/M————

Prices are USD per million tokens; cache read and cache write are the rates for prompt-cache hits and for writing a prompt into the cache. Throughput and latency are measured on AIHubMix — the same figures the model detail page shows — not vendor claims. A dash means the catalog does not publish that field for that model, which is not the same as the model not supporting it.

Common Comparisons

Qwen/QwQ-32B vs Qwen/Qwen2.5-VL-32B-InstructQwen/QwQ-32B vs Qwen/Qwen2.5-VL-72B-InstructQwen/QwQ-32B vs bai-qwen3-vl-235b-a22b-instructQwen/QwQ-32B vs kimi-k3Qwen/QwQ-32B vs glm-5.2

Qwen on AIHubMix

Which Qwen model should I start with?

qwen3-vl-flash at $0.0206/M input — the cheapest entry here that declares tool calling, and it carries a 262K context. Move up to happyhorse-1.0-i2v when answer quality matters more than cost, or to qwen3.6-plus-preview-free for long-form reasoning.

Which of these models reason before answering?

25 of the 149 models here declare a reasoning phase — they work through the problem before producing an answer, which helps on multi-step problems at the cost of extra output tokens. Use the Reasoning filter above the table to see them. The catalog does not record anything further about how they differ, so this page does not sort them into families.

Why are there several entries for the same model?

Because each row is a route you can call, not a model release. Some IDs name an upstream (azure-, alicloud-, cc-), some are the open-weight repository form (Qwen/…), and some differ only in capitalisation, kept so older integrations keep working.

The catalog does not carry a field saying which of those a given row is, so this page does not sort them into buckets it would have to invent. Every row shows that route’s own price, context and speed — compare those directly, and open a model to see the upstreams that serve it.

How is cached input billed?

The Cache read column is the rate for input tokens served from the prompt cache — for example qwen3.5-flash bills cache hits at 10% of the input rate and qwen3.5-plus bills cache hits at 10% of the input rate. Cache write is the surcharge for putting a prompt into the cache in the first place, and only a few upstreams bill it separately. A dash in either column means the catalog carries no cache rate for that model, so plan on paying the full input rate.

Do I need a separate Qwen account?

No. One AIHubMix key covers every model on this page, and switching between them is a change to the model string — billing, rate limits, and logs stay in one place.

Other Model Authors

See all model authors →
OpenAI
142 modelsGoogle
87 modelsZ.AI
71 modelsLlama
53 modelsByteDance
47 models

Start calling Qwen in one line

One key, one endpoint, 911 models across 42 model authors.

Get API keyRead the docs

© 2023 - 2026 AIHubMix, LLC

Product

Developer

Connect

Terms & Policies