llm-benchmarks
githubdrose.io

deepseek-v4.1-flash Benchmarks


Measured on relace

9 runs
Avg Tokens / Second
14.20
Decode Tokens / Second
—
Avg Time to First Token (ms)
5130.00
Runs Analysed
9
Last Updated
Nov 12, 2026, 01:56 PM

Throughput distribution

Throughput over time

deepseek-v4.1-flashrelace250tok/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 · models with at least 4 samples in the window first, fastest of those at the top · shared vertical scale, 0 to 524 tok/s (99th percentile) · dashed rule is the model's own mean

Every lane measuring this model

2 measured · live first · a lane is a provider, a catalogue id and a transport
ProviderLaneAvg Toks/SecMinMaxAvg TTF (ms)n
relacevia OpenRouter14.202.6642.605130.009
relacedirect7.605.399.198680.007

Frequently Asked Questions

How fast is deepseek-v4.1-flash?
The latest rolling average throughput is 14.20 tokens per second with an average time to first token of 5130.00 ms across 9 recent runs.
How often are these benchmarks updated?
Benchmarks refresh automatically whenever the monitoring cron runs. The most recent run completed on Nov 12, 2026, 01:56 PM.

Related models