Summarise · 3 checks · max 400 tokens · benched 2026-09-16

Gemini 3.1 Pro Preview vs Gemini 3.8 Flash vs Gemini 3.5 Flash Lite on meeting notes to exactly three bullets

Google models side by side on "Meeting notes to exactly three bullets": Gemini 3.8 Flash scores 9/10; Gemini 3.8 Flash is the cheapest answer scoring 8+ at $0.34 per 1,000 runs. Outputs, checks, judge reasons, latency and cost.

The prompt every model received

System

You are a concise assistant. Output only what is asked for.

User

Turn these meeting notes into exactly three bullet points. Each bullet starts with "- " and is at most 20 words. Output nothing except the three bullets.

Notes: Launch of the new billing page moves from 18 Sept to 25 Sept because the tax calculation for EU customers still fails for VAT-exempt businesses. Priya owns the fix and will demo it Thursday. Marketing will hold the announcement email until the fix is verified. Separately, the team agreed to drop support for the legacy CSV export in Q1 next year and to notify affected customers in November.

Rubric for the judge: Three bullets that capture the delayed launch, the VAT fix owner, and the CSV export decision; nothing extra.

Side by side

Every cell is one OpenRouter call at temperature 0 with the prompt's token cap and reasoning effort "low" where the model supports it. Cost is usage × the catalogue rate in models.json. Quality is one judge call to anthropic/claude-haiku-4.5 against the prompt's rubric, cached per prompt version.

Gemini 3.1 Pro Preview

google/gemini-3.1-pro-preview

1/10

Latency: 2.2s
Cost: $0.00501
Per 1,000: $5.01

129 in · 396 out (385 reasoning) · 9 words · checks 1/3

regex
max_words
contains

- Billing page launch delayed to Sept 25 due

Judge: Output is incomplete—only one partial bullet provided instead of three complete bullets, failing to meet the core requirement.

Gemini 3.8 Flash

google/gemini-3.8-flash

9/10

top judge scorecheapest scoring 8+

Latency: 918ms
Cost: $0.00034
Per 1,000: $0.34

129 in · 65 out · 53 words · checks 3/3

regex
max_words
contains

- Billing page launch moves to 25 Sept due to an EU VAT calculation bug, which Priya will demo Thursday.
- Marketing is holding the launch announcement email until Priya's tax calculation fix is verified.
- Support for legacy CSV export will end in Q1 next year, with affected customers notified in November.

Judge: All three bullets capture the required points (delayed launch, VAT fix owner, CSV decision) with proper formatting and are under 20 words each, though the first bullet at 19 words is slightly tight on the constraint.

Gemini 3.5 Flash Lite

google/gemini-3.5-flash-lite

1/10

Latency: 975ms
Cost: $0.00103
Per 1,000: $1.03

129 in · 396 out (382 reasoning) · 12 words · checks 1/3

regex
max_words
contains

- The billing page launch is delayed to September 25 due to

Judge: Output is incomplete—only one partial bullet provided instead of three complete bullets as required.

Frequently asked

What does this prompt test?

Summarise: Three bullets that capture the delayed launch, the VAT fix owner, and the CSV export decision; nothing extra. The deterministic checks are regex, max_words, contains.

Which model should I pick for this task?

If the judge's bar of 8/10 is good enough for you, Gemini 3.8 Flash at $0.34 per 1,000 runs. If you need the top score, Gemini 3.8 Flash at $0.34 per 1,000 runs.

Same prompt, other families

Run it yourself

Re-run this prompt on Google with your own OpenRouter key, or tweak the wording and see what changes.