Gemini 3.8 Flash API price, and what each result really costs
Google, listed 2 Sept 2026. Price checked 12 Nov 2026; unchanged since we started tracking it. Source: Google's pricing page. google/gemini-3.8-flash
List price per million tokens
- Input
- $0.75
- Cached input
- $0.075
- Output
- $3.75
- Context window
- 1.05M tokens
Note: Price through 2026-12-31; from 2027-01-01 $1.50 in / $0.15 cached / $7.50 out. Conflict: OpenRouter headline shows $0.825/$4.125 (its +10% regional Google endpoint); official standard is $0.75/$3.75. Cache storage billed extra.
What it costs per task, and per task that works
Six common task shapes with typical token counts. The real cost counts the tries that get redone; the rates are examples, not measurements.
| Task | Per 1,000 tries | Per 1,000 that work, 8 in 10 | Per 1,000 that work, 6 in 10 | Per 1,000 that work, 4 in 10 |
|---|---|---|---|---|
| Support bot answer3,000 in / 300 out | $3.38 | $4.22 | $5.63 | $8.44 |
| Invoice field extraction2,500 in / 250 out | $2.81 | $3.52 | $4.69 | $7.03 |
| Lead scoring1,200 in / 100 out | $1.28 | $1.59 | $2.13 | $3.19 |
| Email reply draft1,500 in / 400 out | $2.63 | $3.28 | $4.38 | $6.56 |
| Translation900 in / 800 out | $3.68 | $4.59 | $6.13 | $9.19 |
| Search re-ranking4,000 in / 50 out | $3.19 | $3.98 | $5.31 | $7.97 |
Work out your own real cost per resultTest Gemini 3.8 Flash on your own prompt
Compare Gemini 3.8 Flash with
- Gemini 3.5 Flash-Lite$0.30 in / $2.50 outSwitch cost
- Gemini 2.5 Flash$0.30 in / $2.50 outSwitch cost
- Grok 4.3$1.25 in / $2.50 outSwitch cost
- GPT-5.4 mini$0.75 in / $4.50 outSwitch cost
- Mistral Large 4$0.70 in / $2.10 outSwitch cost
Worked-through switches: Gemini 2.5 Pro → Gemini 3.8 Flash: cost and quality on real tasks, GPT-4o → Gemini 3.8 Flash: cost and quality on real tasks, Claude Sonnet 5.5 → Gemini 3.8 Flash: cost and quality on real tasks, Gemini 3.8 Flash → Gemini 3.5 Flash-Lite: cost and quality on real tasks
Questions
How much does the Gemini 3.8 Flash API cost?
$0.75 per million input tokens and $3.75 per million output tokens, or $0.075 per million for cached input. Price checked 12 Nov 2026; unchanged since we started tracking it.
What does a result that works really cost with Gemini 3.8 Flash?
A support bot answer (about 3,000 tokens in, 300 out) costs $0.0034 per try. If 6 in 10 work first time, each answer that works really costs $0.0056, because the redone ones were paid for too.