llm-benchmarks
githubdrose.io

DeepSeek V4.1 Flash 1017 Benchmarks


Measured on nebius

16 runs
Avg Tokens / Second
45.70
Decode Tokens / Second
—
Avg Time to First Token (ms)
2210.00
Runs Analysed
16
Last Updated
Nov 22, 2026, 01:06 AM

Throughput distribution

Throughput over time

DeepSeek V4.1 Flash 1017nebius135tok/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 248 tok/s (99th percentile) · dashed rule is the model's own mean

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
nebiusvia OpenRouter45.708.04106.002210.0016

Frequently Asked Questions

How fast is DeepSeek V4.1 Flash 1017?
The latest rolling average throughput is 45.70 tokens per second with an average time to first token of 2210.00 ms across 16 recent runs.
How often are these benchmarks updated?
Benchmarks refresh automatically whenever the monitoring cron runs. The most recent run completed on Nov 22, 2026, 01:06 AM.

Related models