Moonshot AI · Released Jul 16, 2026
Kimi K3 (Low)
Moonshot's reasoning-optimized model featuring verified performance on the OpenRating benchmark index, balanced for deep analytical depth and cost efficiency.
OpenRating Index
17.0
0–100 composite
Reasoning Index
23.0
0–100 composite
Cost / Task
$2.31
benchmark run
Output speed
43 tok/s
blended tok/s
Time to first token
38.0s
median latency
Input price
$3
per 1M tokens
Output price
$15
per 1M tokens
Context window
1M
tokens
Strengths
Intelligence Index score of 17.0 on independent evaluations
39.0 tokens/sec median generation throughput with sub-second response times
1000k token context window supporting complex multi-document reasoning
How we rate Kimi K3 (Low)
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
17.0
Reasoning Index
23.0
Token Pricing
Input per 1M tokens
$3
Output per 1M tokens
$15
Cost of 1M input + 1M output
$18
Task Cost Economics
Empirical cost breakdown for a standard OpenRating Index benchmark task ($2.31 total).
Related models
Compare Kimi K3 (Low) with the models it is most often measured against.