llm-benchmarks
githubdrose.io

kimi-k3-fast Benchmarks


Measured on fireworks

24 runs
fireworks site
Avg Tokens / Second
32.80
Avg Time to First Token (ms)
0.00
Runs Analysed
24
Last Updated
Aug 6, 2026, 09:27 AM

Throughput distribution

  • kimi-k3-fast fireworks33 tok/s
02550

Visible-token throughput, same basis as the table · curves normalised per model

Throughput over time

kimi-k3-fastfireworks1tok/s

Generated throughput, including reasoning tokens — so a thinking model's line does not collapse; the table ranks on visible tokens · one cell per model, showing its best-covered provider · shared vertical scale, 0 to 44 tok/s (99th percentile) · dashed rule is the model's own mean

Every provider serving this model

1 measured
ProviderModelAvg Toks/SecMinMaxAvg TTF (ms)
fireworkskimi-k3-fast32.801.1749.200.00

Frequently Asked Questions

How fast is kimi-k3-fast?
The latest rolling average throughput is 32.80 tokens per second with an average time to first token of 0.00 ms across 24 recent runs.
How often are these benchmarks updated?
Benchmarks refresh automatically whenever the monitoring cron runs. The most recent run completed on Aug 6, 2026, 09:27 AM.