Benchmark Overview
145.00
180.00
41
Apr 16, 2026, 03:01 AM
Key Insights
kimi-k2-instruct streams at 145.00 tokens/second on average across the last 41 benchmark runs.
Performance fluctuated by 132.20 tokens/second (91.2% coefficient of variation), indicating variable behavior across benchmark runs.
Average time to first token is 180.00 ms (excellent latency), suitable for latency-sensitive workloads.
Latest measurements completed on Apr 16, 2026, 03:01 AM based on 41 total samples.
Performance Distribution
Distribution of throughput measurements showing performance consistency across benchmark runs.
Performance Over Time
Historical performance trends showing how throughput has changed over the benchmarking period.
kimi-k2-instruct
Benchmark Samples
| Provider | Model | Avg Toks/Sec | Min | Max | Avg TTF (ms) |
|---|---|---|---|---|---|
| groq | kimi-k2-instruct | 145.00 | 61.80 | 194.00 | 180.00 |
Frequently Asked Questions
The latest rolling average throughput is 145.00 tokens per second with an average time to first token of 180.00 ms across 41 recent runs.
Benchmarks refresh automatically whenever the monitoring cron runs. The most recent run completed on Apr 16, 2026, 03:01 AM.