Gemini 3.8 Flash API Pricing: $0.75 / $3.75 per 1M Tokens
The numbers: Gemini 3.8 Flash (Google DeepMind) bills $0.75 per 1M input tokens, $3.75 per 1M output tokens, and $0.075 per 1M cached input tokens, inside a 1M context window. On our reference 200-step agent workload that works out to about $0.97 per task.
Newest Flash model for long-horizon software engineering and autonomous agents; 1M context, 64K output. Every figure on this page was read directly from Google DeepMind’s own documentation on September 18, 2026 — not copied from a third-party comparison table. Rates change; the source link and check date below are part of the data.
Verified rates
| Item | Rate | Unit |
|---|---|---|
| Input | $0.75 | per 1M tokens |
| Output | $3.75 | per 1M tokens |
| Cached input (cache read) | $0.075 | per 1M tokens |
| Context window | 1M | tokens |
Billing notes: Launched September 2, 2026 on the same introductory rate as Gemini 3.6 and 3.7 Flash. Standard pricing of $1.50 / $7.50 per 1M applies from January 1, 2027. Thinking tokens bill at the output rate.
Source: Google DeepMind official pricing ↗, checked September 18, 2026.
What a real workload costs on Gemini 3.8 Flash
List rates are not bills. These two reference workloads translate the table above into money. The assumptions are visible so you can swap in your own trace — the cost calculator does exactly that.
Scenario A — one 200-step agent task
A coding agent runs 200 steps against a stable 40,000-token prefix (system prompt, tools, repository map). The prefix is written to cache on the first step, re-read on the remaining 199, and each step produces 500 output tokens. For Google DeepMind, cache writes bill at the input rate (per-hour cache storage is excluded).
| Component | Calculation | Cost |
|---|---|---|
| Cache write (once) | billed via hourly cache storage (excluded) | — |
| Cache reads | 199 × 40,000 × $0.075 / 1M | $0.60 |
| Output | 200 × 500 × $3.75 / 1M | $0.38 |
| Total per task | $0.97 |
Scenario B — one heavy month
A production service pushes 50M fresh input tokens, 150M cached input tokens, and 10M output tokens through Gemini 3.8 Flash in a month:
| Component | Volume | Cost |
|---|---|---|
| Fresh input | 50M × $0.75 | $37.50 |
| Cached input | 150M × $0.075 | $11.25 |
| Output | 10M × $3.75 | $37.50 |
| Monthly total | $86.25 |
Head-to-head cost comparisons
- GPT-5.6 Luna vs Gemini 3.8 Flash — same workload, $0.29 vs $0.97 (3.4x apart).
- Gemini 3.8 Flash vs DeepSeek V4.1 Flash — same workload, $0.97 vs $0.09 (10.8x apart).
- Claude Haiku 4.5 vs Gemini 3.8 Flash — same workload, $1.35 vs $0.97 (1.4x apart).
Frequently asked questions
How much does Gemini 3.8 Flash cost per million tokens?
Gemini 3.8 Flash costs $0.75 per 1M input tokens and $3.75 per 1M output tokens, with cached input at $0.075 per 1M. Verified September 18, 2026 against Google DeepMind's official pricing.
What is the context window of Gemini 3.8 Flash?
Gemini 3.8 Flash supports a 1M token context window, per Google DeepMind's documentation checked September 18, 2026.
Does Gemini 3.8 Flash support prompt caching?
Yes. Cached input tokens bill at $0.075 per 1M instead of the full input rate; cache writes bill at the input rate (per-hour cache storage is excluded).
When was this Gemini 3.8 Flash price last checked?
On September 18, 2026, against Google DeepMind's own pricing page: https://deepmind.google/models/gemini/flash/