Benchmark Overview
50.60
0.81
869
Oct 18, 2025, 11:00 PM
Key Insights
llama-3.1-405b streams at 50.60 tokens/second on average across the last 869 benchmark runs.
Performance fluctuated by 69.10 tokens/second (136.6% coefficient of variation), indicating variable behavior across benchmark runs.
Average time to first token is 0.81 ms (excellent latency), suitable for latency-sensitive workloads.
Latest measurements completed on Oct 18, 2025, 11:00 PM based on 869 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.
llama-3.1-405b
Benchmark Samples
| Provider | Model | Avg Toks/Sec | Min | Max | Avg TTF (ms) |
|---|---|---|---|---|---|
| together | llama-3.1-405b | 50.60 | 1.60 | 70.70 | 0.81 |
Frequently Asked Questions
The latest rolling average throughput is 50.60 tokens per second with an average time to first token of 0.81 ms across 869 recent runs.
Benchmarks refresh automatically whenever the monitoring cron runs. The most recent run completed on Oct 18, 2025, 11:00 PM.