Today's price · per 1M tokens
Input
$0.750
per 1M tokens
Output
$3.75
per 1M tokens
Blended
$1.50
blended $/1M — 3:1 weighted input:output
Cached input
$0.075
10% of input — prompt caching
How this price is scoped: Google's Gemini API pricing page prints a dated, two-phase schedule for this model: $0.75 per 1M input tokens through December 31, 2026, rising to $1.50 on January 1, 2027, and $3.75 per 1M output tokens through December 31, 2026, rising to $7.50. Context caching follows the same dates ($0.075, then $0.15). We publish the rate you would pay today and re-verify it before the changeover date. Note that resellers may quote a lower figure: OpenRouter's ':batch' route for this model shows $0.375/$1.875, which is Google's own Batch tier — exactly half the standard rate, and not what an ordinary API call is billed. We publish the standard rate. Checked 2026-09-03 against ai.google.dev/gemini-api/docs/pricing, in the section headed 'Gemini 3.8 Flash' (gemini-3.8-flash).
Price receipt
We read Google's own pricing page on and found Gemini 3.8 Flash listed with a price on the same card — that read is where the number above comes from. We keep the page text we read; its content hash is 60a63d484d.
Overview
Google's most intelligent Flash model, launched 2026-09-02 as the successor to Gemini 3.7 Flash — engineered for long-horizon software engineering, autonomous agents and complex enterprise workflows. Currently $0.75/1M input and $3.75/1M output on Google's dated schedule, reverting to $1.50/$7.50 on 2027-01-01. 1M-token context with native image, audio and video input.
Capabilities
struck through = not supportedBenchmark performance
accuracy % · higher is better- AA Intelligence Index: 40.9 (Artificial Analysis composite across reasoning, knowledge and coding evals)
- AA-Omniscience Index: 29.6 (−100–100)
- GDPval-AA v2: 1463.8 ELO (Real-world work tasks, human baseline = 1000)
- LMSYS Chatbot Arena: 1493 ELO (Human preference)
- Ranks #10 of 100 benchmarked models by average score
- Ranks #22 of 195 comparably-measured models by percentile score, across 9 independent measurements
- Ranks #54 of 195 comparably-tested models by normalized performance per dollar
- Strongest at Terminal-Bench 2.1 — 89.4%, #2 of 48
277.5 tokens/sec output
Source: Vellum LLM Leaderboard · updated · See full rankings →
Specifications
- Provider
- Context window
- 1M tokens
- Modality
- text, image, audio, video
- Parameters
- Proprietary
- Open source
- No — proprietary
- Released
- Status
- Current
- Last updated
- Tags
Availability verified: — listed on Google's own page