Moonshot AI · Released May 20, 2026
Kimi K3 (max)
Moonshot AI's frontier open-weights reasoning model with explicit multi-step thinking traces and deep STEM problem solving.
OpenRating Index
59.8
0–100 composite
Reasoning Index
85.5
0–100 composite
Cost / Task
$0.84
benchmark run
Output speed
38 tok/s
blended tok/s
Time to first token
1.3s
median latency
Input price
$0.9
per 1M tokens
Output price
$3.6
per 1M tokens
Context window
256K
tokens
Strengths
Explicit chain-of-thought output with granular reasoning
Open weights suitable for custom inference deployment
Exceptional long-form synthesis and bilingual mastery
How we rate Kimi K3 (max)
OpenRating scores every model on the same axes. The OpenRating Index is a weighted composite of reasoning, knowledge, and agentic benchmarks, normalised to a 0–100 scale. Speed is blended output measured in tokens per second under standardised load, with time-to-first-token reported separately. Prices are list input and output prices per million tokens, and Task Cost measures end-to-end workload spending.
OpenRating Index
59.8
Reasoning Index
85.5
Token Pricing
Input per 1M tokens
$0.9
Output per 1M tokens
$3.6
Cost of 1M input + 1M output
$4.5
Task Cost Economics
Empirical cost breakdown for a standard OpenRating Index benchmark task ($0.84 total).
Related models
Compare Kimi K3 (max) with the models it is most often measured against.