llm-benchmarks
githubdrose.io

Kimi K2.6 Benchmarks


Measured on gmicloud

176 runs
Avg Tokens / Second
1.65
Avg Time to First Token (ms)
36300.00
Runs Analysed
176
Last Updated
Sep 1, 2026, 03:12 PM

Throughput distribution

Throughput over time

Kimi K2.6coreweave4tok/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 200 tok/s (99th percentile) · dashed rule is the model's own mean · 27 further providers not drawn

Every provider serving this model

1 measured
ProviderModelAvg Toks/SecMinMaxAvg TTF (ms)
gmicloudKimi K2.61.651.651.6536300.00

Frequently Asked Questions

How fast is Kimi K2.6?
The latest rolling average throughput is 1.65 tokens per second with an average time to first token of 36300.00 ms across 176 recent runs.
How often are these benchmarks updated?
Benchmarks refresh automatically whenever the monitoring cron runs. The most recent run completed on Sep 1, 2026, 03:12 PM.

Related models