← LLM 
Baseten
DeepSeek-V4.1-Flash
Measured 2026-10-04
Run DeepSeek-V4.1-Flash on Speko →One API, every model on this board.
Score
81
Task done
82.19%
69–94% 95% CI bootstrap over items · n=95 · measured 2026-10-04
Refusal quality
74.4%
names the gap 70% · offers a route 60% · 10 probes x 3 iterations · 1 silent turn scored zero
Fabrication
0%
0 of 10 probes (30 probe runs) · ≤31%
Dead-air
36.6%
117 of 320 turns
Tool silence
52.7%
117 of 222 tool-call turns
Stalled
16.7%
15 of 90 runs
TTFT p50
458ms
458–1062 p50–p90 · n=38
Cost / 1M tok
$1.20
Share this result
Embed the live badge
Markdown
[](https://benchmarks.speko.ai/llm/baseten-deepseek-v4-1-flash)
HTML
<a href="https://benchmarks.speko.ai/llm/baseten-deepseek-v4-1-flash"><img src="https://benchmarks.speko.ai/badge/llm/baseten-deepseek-v4-1-flash.svg" alt="Speko llm rank"></a>
URL
https://benchmarks.speko.ai/llm/baseten-deepseek-v4-1-flash