Gemini 3.8 Flash Listed: $0.75 Input, 1049k Context, Multi-Modal

Google Adds Gemini 3.8 Flash to the Roster

Google listed Gemini 3.8 Flash on 2026-09-02, slotting it into the mid-tier Flash family with a notably wide context window and multi-modal input support.

Pricing and Context at a Glance

• Input: $0.75 per 1M tokens
• Output: $3.75 per 1M tokens
• Context window: 1049k tokens

The 5:1 output-to-input price ratio is fairly standard for Flash-class models, but the 1049k context window is worth flagging — it's large enough to handle lengthy codebases, legal documents, or extended conversation histories in a single pass without chunking workarounds.

Modalities

The model accepts text, image, video, file, and audio inputs and returns text. That breadth makes it relevant for transcription pipelines, document Q&A, and mixed-media workflows where swapping in a specialist model per modality would add latency and cost.

Who Should Care

Developers running long-context tasks get a competitive window without moving to a full "Pro" tier price. At $0.75 input, processing a 1M-token document costs $0.75 — manageable for batch jobs, tighter for real-time applications where output volume drives the bill.

Teams already on Gemini 2.x Flash should compare throughput and quality on their actual workloads before migrating; pricing alone doesn't tell the whole story.

Users of competing mid-range models (Claude Haiku, GPT-4o Mini, etc.) now have another data point — check our comparison tables to see where 3.8 Flash lands on price-per-output-token across the board.

Full pricing details, availability status, and historical price changes are tracked on the Gemini 3.8 Flash — live specs & price history page.

Ezra, Scout AI Team