Moonshot AI

Kimi K3

Moonshot AI's frontier model (Kimi K3, 1M context). Newly added — Playground benchmark metrics are still being gathered. The first Moonshot model in the lineup.

Avg tokens per run: 99,260
Avg duration: 79.8s
Turns to checkout: 12

Start a run with Kimi K3 Compare

Shopping score

15/100Weak

Scores come from runs on UCP stores, not estimates: every figure is a tool call, a checkout attempt or a store’s answer.

Shopping score breakdown

Checkout rate40%3%
Cart rate20%33.3%
Search rate10%72.7%
Turn efficiency15%0%
Token efficiency15%0%

24.2% of runs failed with errors.

Top stores

Development stores are hidden.

Store	Checkout	Cart
allbirds.com	0%	33.3%
store hidden	0%	100%
store hidden	0%	0%
unionmadestickers.com	0%	33.3%
store hidden	0%	0%

Known issues

LowMetrics Pending
Kimi K3 was recently added; its Playground benchmark metrics are still being gathered.

Token usage

Avg per run: 99,260
Average of all models: 77,803
Against the average: +28%
Median (p50): 57,691
p90: 208,663

Prompt and completion

Prompt 97,170 (98%)Completion 2,090 (2%)

Daily average tokens per run, last 30 days

Oct 13Total 11,152Prompt 10,753Completion 399

Range

Min: 2,796
p25: 22,619
p50: 57,691
p75: 89,645
p90: 208,663
Max: 978,058

Test Kimi K3 on your store

Start a run to see how Kimi K3 handles your store’s checkout, call by call.

Start a run