Kimi K3
Kimi K3 is a model from Moonshot AI, run through Amazon Bedrock. It places 1= of 14 on the Hurdle Index, and clears the 90% bar on pay-gap reasoning.
- Lab
- Moonshot AI
- Called as
us.moonshotai.kimi-k3- Price, per million tokens
- $3.3 in, $16.5 out
- Tokens per task
- 1,015 read, 19,513 written
- Output limit
- 32,000 tokens
- Run
- 23 October 2026, 120 attempts
Hurdle Index
98.7%96.7% to 100.0%. Place 1= of 14.Cost per task
$0.32514th cheapest of 14. List price of the tokens used.Time per task
130 s13th fastest of 14. Wall-clock time of each call.On each benchmark
| Benchmark | Mean, 95% interval | Place | Every run | Best of 3 | Cost per task | Cut off | Bar |
|---|---|---|---|---|---|---|---|
| Pay-gap reasoning | 98.7% 96.7% to 100.0% | 1= of 14 | 96.1% | 100.0% | $0.325 | none | Clears |
Every attempt
Pay-gap reasoning: 40 tasks, three attempts each, one cell per attempt. White is a perfect answer, black is none of it right.
Attempt 1
Attempt 2
Attempt 3
Where it misses
The share of each kind of rubric line it got wrong, across every attempt.
Next to it on the index
Is it right for your job?
A public benchmark says how it does on someone else’s work. Hurdle builds one from yours.
Book a walkthrough