llm-benchmarks
githubdrose.io

deepseek-v4.1-flash Benchmarks


Measured on primeintellect

1 runs
Avg Tokens / Second
3.97
Decode Tokens / Second
—
Avg Time to First Token (ms)
15800.00
Runs Analysed
1
Last Updated
Nov 10, 2026, 07:32 PM

Throughput distribution

Throughput over time

deepseek-v4.1-flashprimeintellect93tok/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 98 tok/s (99th percentile) · dashed rule is the model's own mean · 1 cell carry fewer than 4 samples

Every lane measuring this model

1 measured · live first · a lane is a provider, a catalogue id and a transport
ProviderLaneAvg Toks/SecMinMaxAvg TTF (ms)n
primeintellectdirect3.973.973.9715800.001

Frequently Asked Questions

How fast is deepseek-v4.1-flash?
The latest rolling average throughput is 3.97 tokens per second with an average time to first token of 15800.00 ms across 1 recent runs.
How often are these benchmarks updated?
Benchmarks refresh automatically whenever the monitoring cron runs. The most recent run completed on Nov 10, 2026, 07:32 PM.

Related models