llm-benchmarks
githubdrose.io

gpt-3.5-turbo Benchmarks


Measured on openai

1990 runs
openai site
Avg Tokens / Second
41.50
Avg Time to First Token (ms)
1220.00
Runs Analysed
1990
Last Updated
Aug 10, 2026, 06:49 AM

Throughput distribution

  • gpt-3.5-turbo openai42 tok/s
0255075

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

Throughput over time

gpt-3.5-turboopenai29tok/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 77 tok/s (99th percentile) · dashed rule is the model's own mean

Every provider serving this model

1 measured
ProviderModelAvg Toks/SecMinMaxAvg TTF (ms)
openaigpt-3.5-turbo41.503.6878.101220.00

Frequently Asked Questions

How fast is gpt-3.5-turbo?
The latest rolling average throughput is 41.50 tokens per second with an average time to first token of 1220.00 ms across 1990 recent runs.
How often are these benchmarks updated?
Benchmarks refresh automatically whenever the monitoring cron runs. The most recent run completed on Aug 10, 2026, 06:49 AM.

Related models