← LLM

Gemini

gemini-3.5-flash-lite

reasoning at minimal (family floor)

Measured 2026-08-25

* Conditional

Measured 2026-07-25; drops some compound multi-step requests.

Score

73

Task done

72.02%

53–89% 95% CI bootstrap over items · n=95 · measured 2026-08-25

Refusal quality

70.0%

names the gap 63% · offers a route 70% · 10 probes x 3 iterations · 6 silent turns scored zero

Fabrication

10%

3 of 30 probe runs

Dead-air

68.6%

197 of 287 turns

Tool silence

100.0%

all of 19 items (197 tool-call turns) · ≥82%

Stalled

27.8%

25 of 90 runs

TTFT p50

373ms

373–404 p50–p90

Cost / 1M tok

$2.50

Share this result

Embed the live badge

Speko llm rank
Markdown
[![Speko llm rank](https://benchmarks.speko.ai/badge/llm/gemini-gemini-3-5-flash-lite.svg)](https://benchmarks.speko.ai/llm/gemini-gemini-3-5-flash-lite)
HTML
<a href="https://benchmarks.speko.ai/llm/gemini-gemini-3-5-flash-lite"><img src="https://benchmarks.speko.ai/badge/llm/gemini-gemini-3-5-flash-lite.svg" alt="Speko llm rank"></a>
URL
https://benchmarks.speko.ai/llm/gemini-gemini-3-5-flash-lite