Moonshot AI Models

35 modelsGeneral models free to startUp to 1.05M context

Usage

Last 30 days · 2026-10-21 to 2026-11-19

Tokens

89.7B

Requests

319K

Models in use

31 of 35

Tokens per day, stacked by model

03.4B6.8B10-2110-2811-0411-1111-182026-10-21 — 1,343,576,425 tokens kimi-k3: 1,188,443,640 kimi-k2.6: 129,972,915 coding-kimi-k3-free: 16,864,980 8 more models: 4,152,310 kimi-k2.7-code: 4,011,940 coding-kimi-k3: 130,6402026-10-22 — 1,361,449,540 tokens kimi-k3: 1,118,671,180 kimi-k2.6: 125,117,640 coding-kimi-k3-free: 84,517,555 8 more models: 17,714,080 kimi-k2.7-code: 12,800,875 coding-kimi-k3: 2,628,2102026-10-23 — 1,437,981,435 tokens kimi-k3: 1,144,922,935 coding-kimi-k3-free: 173,910,920 kimi-k2.6: 107,689,255 8 more models: 7,798,250 kimi-k2.7-code: 1,932,190 coding-kimi-k3: 1,727,8852026-10-24 — 1,099,441,705 tokens kimi-k3.1-pro: 822,799,350 coding-kimi-k3-free: 116,854,830 kimi-k3: 88,038,720 8 more models: 32,314,005 kimi-k2.6: 22,690,580 coding-kimi-k3: 11,368,025 coding-kimi-k3.1-pro: 3,374,055 kimi-k3.1-pro-highspeed: 1,789,635 kimi-k2.7-code: 212,5052026-10-25 — 877,071,735 tokens kimi-k3.1-pro: 616,093,040 8 more models: 104,586,585 kimi-k3: 79,475,465 coding-kimi-k3-free: 32,345,040 kimi-k2.7-code: 26,252,245 coding-kimi-k3.1-pro: 7,964,235 kimi-k2.6: 6,507,615 kimi-k3.1-pro-highspeed: 3,500,860 coding-kimi-k3: 346,6502026-10-26 — 975,111,080 tokens kimi-k3.1-pro: 625,475,905 kimi-k3: 120,710,675 coding-kimi-k3-free: 110,856,520 kimi-k3.1-pro-highspeed: 97,793,015 coding-kimi-k3: 5,543,100 kimi-k2.7-code: 5,231,285 8 more models: 4,768,195 kimi-k2.6: 4,726,965 coding-kimi-k3.1-pro: 5,4202026-10-27 — 2,228,906,525 tokens kimi-k3.1-pro-highspeed: 950,680,335 kimi-k3.1-pro: 712,055,720 kimi-k3: 397,730,175 coding-kimi-k3-free: 123,167,145 kimi-k2.7-code: 20,336,715 8 more models: 13,025,530 coding-kimi-k3: 7,723,315 kimi-k2.6: 4,184,885 coding-kimi-k3.1-pro: 2,7052026-10-28 — 2,952,737,645 tokens kimi-k3.1-pro: 1,747,374,220 kimi-k3.1-pro-highspeed: 784,769,090 kimi-k3: 199,922,645 coding-kimi-k3-free: 149,126,870 kimi-k2.7-code: 51,251,945 8 more models: 11,019,100 kimi-k2.6: 8,512,905 coding-kimi-k3: 758,135 coding-kimi-k3.1-pro: 2,7352026-10-29 — 3,828,688,425 tokens kimi-k3.1-pro: 1,990,403,135 kimi-k3.1-pro-highspeed: 1,599,708,545 coding-kimi-k3-free: 88,602,010 kimi-k3: 56,520,130 coding-kimi-k3.1-pro: 29,102,400 8 more models: 28,693,405 kimi-k2.6: 18,217,160 kimi-k2.7-code: 16,434,605 coding-kimi-k3: 1,007,0352026-10-30 — 3,218,239,580 tokens kimi-k3.1-pro: 2,592,900,965 kimi-k3: 233,763,575 kimi-k3.1-pro-highspeed: 169,200,825 coding-kimi-k3-free: 107,711,975 kimi-k2.6: 40,310,090 kimi-k2.7-code: 36,351,720 8 more models: 19,845,275 coding-kimi-k3: 18,154,295 coding-kimi-k3.1-pro: 8602026-10-31 — 1,544,336,165 tokens kimi-k3.1-pro-highspeed: 1,063,089,980 kimi-k3.1-pro: 225,283,260 coding-kimi-k3-free: 97,318,440 kimi-k2.7-code: 57,512,060 kimi-k3: 48,518,325 kimi-k2.6: 20,473,660 8 more models: 16,677,940 coding-kimi-k3: 15,460,955 coding-kimi-k3.1-pro: 1,5452026-11-01 — 1,692,062,575 tokens kimi-k3.1-pro: 1,248,598,920 kimi-k3: 157,691,560 kimi-k2.7-code: 94,832,560 coding-kimi-k3-free: 83,962,775 coding-kimi-k3.1-pro: 53,305,565 8 more models: 24,418,690 coding-kimi-k3: 19,270,195 kimi-k3.1-pro-highspeed: 9,122,860 kimi-k2.6: 859,4502026-11-02 — 5,678,130,630 tokens kimi-k3.1-pro: 4,631,350,490 kimi-k3.1-pro-highspeed: 489,313,055 kimi-k3: 194,462,040 kimi-k2.6: 152,589,115 kimi-k2.7-code: 122,061,825 coding-kimi-k3-free: 78,173,635 8 more models: 9,434,895 coding-kimi-k3: 743,650 coding-kimi-k3.1-pro: 1,9252026-11-03 — 4,803,247,890 tokens kimi-k3.1-pro: 4,226,257,000 kimi-k2.7-code: 269,752,505 coding-kimi-k3-free: 99,874,135 kimi-k3: 95,707,795 kimi-k2.6: 83,513,305 coding-kimi-k3: 14,766,430 kimi-k3.1-pro-highspeed: 11,160,995 8 more models: 2,215,7252026-11-04 — 3,193,148,960 tokens kimi-k3.1-pro: 2,852,436,445 coding-kimi-k3-free: 106,727,505 kimi-k2.7-code: 89,889,060 kimi-k3.1-pro-highspeed: 85,041,950 kimi-k2.6: 30,836,310 kimi-k3: 15,787,295 8 more models: 11,690,980 coding-kimi-k3: 726,095 coding-kimi-k3.1-pro: 13,3202026-11-05 — 5,356,992,665 tokens kimi-k3.1-pro: 4,553,010,710 kimi-k3: 612,277,175 coding-kimi-k3-free: 102,439,565 kimi-k2.7-code: 39,279,970 8 more models: 16,099,775 kimi-k2.6: 15,570,275 kimi-k3.1-pro-highspeed: 13,380,405 coding-kimi-k3: 4,933,975 coding-kimi-k3.1-pro: 8152026-11-06 — 6,762,277,745 tokens kimi-k3: 3,620,337,590 kimi-k3.1-pro: 2,833,872,635 coding-kimi-k3-free: 135,219,960 kimi-k2.6: 76,057,885 kimi-k2.7-code: 40,400,720 8 more models: 36,307,095 kimi-k3.1-pro-highspeed: 12,443,675 coding-kimi-k3.1-pro: 5,122,155 coding-kimi-k3: 2,516,0302026-11-07 — 5,379,102,900 tokens kimi-k3: 4,174,181,445 kimi-k3.1-pro: 854,472,635 coding-kimi-k3-free: 136,316,430 coding-kimi-k3: 116,372,645 coding-kimi-k3.1-pro: 70,966,865 kimi-k2.7-code: 9,860,045 8 more models: 8,896,660 kimi-k2.6: 7,395,265 kimi-k3.1-pro-highspeed: 640,9102026-11-08 — 5,023,556,805 tokens kimi-k3: 3,334,159,990 kimi-k3.1-pro: 1,539,911,640 coding-kimi-k3-free: 82,598,440 8 more models: 29,613,240 kimi-k2.6: 15,812,545 coding-kimi-k3.1-pro: 13,652,850 kimi-k2.7-code: 3,884,860 coding-kimi-k3: 3,456,400 kimi-k3.1-pro-highspeed: 466,8402026-11-09 — 5,061,004,070 tokens kimi-k3.1-pro: 3,702,020,555 kimi-k3: 1,049,390,920 kimi-k2.6: 113,376,360 coding-kimi-k3-free: 84,067,895 coding-kimi-k3: 57,981,835 coding-kimi-k3.1-pro: 18,293,460 kimi-k3.1-pro-highspeed: 15,948,945 8 more models: 13,516,985 kimi-k2.7-code: 6,407,1152026-11-10 — 3,000,218,515 tokens kimi-k3.1-pro: 2,245,391,045 kimi-k3: 532,557,015 coding-kimi-k3.1-pro: 72,159,435 coding-kimi-k3-free: 60,427,120 kimi-k2.6: 31,269,765 kimi-k2.7-code: 30,366,475 8 more models: 15,692,700 kimi-k3.1-pro-highspeed: 11,707,770 coding-kimi-k3: 647,1902026-11-11 — 3,140,324,925 tokens kimi-k3.1-pro: 2,193,802,595 kimi-k3: 759,078,740 coding-kimi-k3-free: 71,562,560 kimi-k2.7-code: 41,235,010 coding-kimi-k3.1-pro: 38,359,705 kimi-k2.6: 23,027,535 8 more models: 12,906,350 coding-kimi-k3: 267,540 kimi-k3.1-pro-highspeed: 84,8902026-11-12 — 3,396,364,825 tokens kimi-k3.1-pro: 2,655,804,430 coding-kimi-k3.1-pro: 476,794,740 coding-kimi-k3-free: 111,125,395 kimi-k2.7-code: 80,538,025 kimi-k3.1-pro-highspeed: 36,715,035 kimi-k3: 18,795,390 kimi-k2.6: 7,838,980 8 more models: 6,191,455 coding-kimi-k3: 2,561,3752026-11-13 — 2,415,864,675 tokens kimi-k3.1-pro: 2,084,239,015 coding-kimi-k3.1-pro: 134,139,885 kimi-k2.6: 66,970,570 coding-kimi-k3-free: 65,623,355 kimi-k2.7-code: 30,123,115 kimi-k3: 17,293,945 8 more models: 14,461,620 coding-kimi-k3: 1,636,895 kimi-k3.1-pro-highspeed: 1,376,2752026-11-14 — 1,137,863,630 tokens kimi-k3.1-pro: 669,662,355 coding-kimi-k3-free: 143,172,640 coding-kimi-k3.1-pro: 134,550,495 kimi-k2.6: 88,039,630 kimi-k3: 64,184,930 kimi-k2.7-code: 27,056,240 8 more models: 8,310,920 coding-kimi-k3: 2,862,385 kimi-k3.1-pro-highspeed: 24,0352026-11-15 — 3,741,591,715 tokens kimi-k3.1-pro: 3,209,235,045 kimi-k2.6: 244,474,180 kimi-k3.1-pro-highspeed: 192,247,285 kimi-k2.7-code: 55,813,260 kimi-k3: 36,127,475 coding-kimi-k3.1-pro: 3,334,685 coding-kimi-k3-free: 210,345 8 more models: 149,4402026-11-16 — 1,709,439,180 tokens kimi-k3.1-pro: 1,513,121,055 kimi-k2.7-code: 98,226,760 kimi-k2.6: 69,517,895 kimi-k3: 27,056,705 8 more models: 945,735 kimi-k3.1-pro-highspeed: 566,225 coding-kimi-k3-free: 2,555 coding-kimi-k3: 2,2502026-11-17 — 1,781,460,780 tokens kimi-k3.1-pro: 1,612,579,330 kimi-k3: 52,760,240 kimi-k2.7-code: 44,730,025 kimi-k2.6: 32,165,200 coding-kimi-k3-free: 31,134,425 8 more models: 7,283,285 coding-kimi-k3: 659,280 kimi-k3.1-pro-highspeed: 146,955 coding-kimi-k3.1-pro: 2,0402026-11-18 — 2,167,789,145 tokens kimi-k3.1-pro: 1,835,639,255 coding-kimi-k3-free: 124,780,970 kimi-k2.7-code: 75,532,270 kimi-k3: 70,878,960 kimi-k3.1-pro-highspeed: 25,106,560 coding-kimi-k3.1-pro: 15,014,250 coding-kimi-k3: 9,970,185 kimi-k2.6: 6,083,745 8 more models: 4,782,9502026-11-19 — 3,402,112,225 tokens kimi-k3.1-pro: 2,856,350,705 coding-kimi-k3: 228,794,910 coding-kimi-k3.1-pro: 132,930,335 coding-kimi-k3-free: 87,435,525 kimi-k2.7-code: 46,678,095 kimi-k3: 36,716,190 8 more models: 7,961,255 kimi-k2.6: 3,459,335 kimi-k3.1-pro-highspeed: 1,785,875
  • kimi-k3.1-pro
  • kimi-k3
  • kimi-k3.1-pro-highspeed
  • coding-kimi-k3-free
  • kimi-k2.6
  • kimi-k2.7-code
  • coding-kimi-k3.1-pro
  • coding-kimi-k3
  • 8 more models

Which models that traffic went to

  1. Kimi K3.1 Pro63.1%56.7B
  2. Kimi K321.8%19.5B
  3. Kimi K3.1 Pro Highspeed6.2%5.6B
  4. Coding Kimi K3 (free)3.0%2.7B
  5. Kimi K2.61.7%1.6B
  6. Kimi K2.7 Code1.6%1.4B
  7. Coding Kimi K3.1 Pro1.3%1.2B
  8. Coding Kimi K30.6%533M
  9. 8 more models0.5%491M

Share of 89.7B tokens. 15 models with traffic report no token counts and cannot be ranked here, including kimi-k2-instruct and cc-kimi-dev-72b — 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 35 model IDs listed on this page; traffic routed through upstream-specific IDs that are not in the public catalog is not included.

All 35 Moonshot AI Models

Open in model list
Moonshot AI 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
coding-kimi-k3-freeTakes text, vision, video, returns text.1.05M—FreeFree/MFree/M28 tok/s17.21 s
coding-kimi-k3Takes text, vision, video, returns text.1.05M—$0.44$1.6133/M$0.066/M24 tok/s9.51 s
kimi-k3.1-proTakes text, vision, video, returns text.1.05M—$2$10/M$0.2/M61 tok/s1.39 s
kimi-k3Takes text, vision, video, returns text.1.05M—$2.84$14.2/M$0.284/M73 tok/s0.91 s
kimi-k3.1-pro-highspeedTakes text, vision, video, returns text.1.05M—$4$21/M$0.2/M38 tok/s4.26 s
k2.6-code-preview-freeTakes text, vision, video, returns text.262K256KFreeFree/M—30 tok/s13.95 s
cc-k2.6-code-previewTakes text, vision, video, returns text.262K—$0.2$0.2/M$0.02/M3 tok/s10.26 s
kimi-k2-thinkingTakes text, returns text.262K—$0.548$2.192/M$0.137/M——
kimi-k2.5Takes text, vision, video, returns text.262K—$0.6$3/M$0.105/M80 tok/s0.95 s
kimi-k2.6Takes text, vision, video, returns text.262K33K$0.95$3.9995/M$0.1608/M56 tok/s0.98 s
kimi-k2.7-codeTakes text, vision, video, returns text.262K33K$0.95$3.9995/M$0.1608/M43 tok/s1.02 s
kimi-k2-turbo-previewTakes text, returns text.262K8K$1.2$4.8/M$0.3/M20 tok/s3.06 s
kimi-k2.7-code-highspeedTakes text, vision, video, returns text.262K33K$1.9$7.999/M$0.3217/M18 tok/s9.51 s
kimi-for-coding-freeTakes text, vision, video, returns text.256K33KFreeFree/M—33 tok/s1.84 s
sf-kimi-k2-thinkingTakes , returns text.256K—$0.548$2.192/M—42 tok/s1.63 s
kimi-k2-0711Takes text, returns text.131K16K$0.54$2.16/M—50 tok/s0.78 s
cc-kimi-for-codingTakes , returns text.——$0.2$0.2/M$0.02/M25 tok/s7.86 s
moonshotai/Moonlight-16B-A3B-InstructTakes , returns text.——$0.2$0.2/M———
cc-kimi-dev-72bTakes , returns text.——$0.32$1.28/M———
moonshotai/Kimi-Dev-72BTakes , returns text.——$0.32$1.28/M———
kimi-k2-instructTakes text, returns text.——$0.54$2.16/M—40 tok/s0.10 s
alicloud-kimi-k2-instructTakes , returns text.——$0.548$2.192/M—1 tok/s1.23 s
alicloud-kimi-k2-thinkingTakes , returns text.——$0.548$2.192/M———
cc-kimi-k2-thinkingTakes , returns text.——$0.548$2.192/M———
coding-kimi-k3.1-proTakes text, returns text.——$0.88$3.23/M$0.132/M25 tok/s10.80 s
cc-kimi-k2-instructTakes text, returns text.——$1.1$3.3/M———
cc-kimi-k2-instruct-0905Takes text, returns text.——$1.1$3.3/M———
coding-kimi-k3.1-pro-highspeedTakes text, vision, returns text.——$1.76$6.46/M$0.264/M18 tok/s1.76 s
moonshot-v1-8kTakes , returns text.——$2$2/M———
moonshot-v1-8k-vision-previewTakes , returns text.——$2$2/M———
moonshot-v1-32kTakes , returns text.——$4$4/M———
moonshot-v1-32k-vision-previewTakes , returns text.——$4$4/M———
moonshot-v1-128kTakes , returns text.——$10$10/M———
moonshot-v1-128k-vision-previewTakes , returns text.——$10$10/M———
kimi-thinking-previewTakes , returns text.——$30$30/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.

Moonshot AI on AIHubMix

Which Moonshot AI model should I start with?

coding-kimi-k3-free is free on input — the cheapest entry here that declares tool calling, and it carries a 1.05M context. Move up to kimi-thinking-preview when answer quality matters more than cost, or to coding-kimi-k3 for long-form reasoning.

Which of these models reason before answering?

10 of the 35 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 (moonshotai/…), 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 kimi-k3.1-pro-highspeed bills cache hits at 5% of the input rate and cc-k2.6-code-preview 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 Moonshot AI 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.

Start calling Moonshot AI in one line

One key, one endpoint, 983 models across 49 model authors.