Google · Gemini
Gemini 3.8 Flash
Google's current fastest Flash model for complex agentic tasks at scale, positioned as the successor to 3.7 Flash with stronger long-horizon software engineering and multi-step reasoning, with its own current API price published in Google’s pricing table.
Start here
- What this helps you do
- Decide whether Gemini 3.8 Flash is the right model for a job you have, or one to leave alone.
- What you leave with
- A plain read of what Gemini 3.8 Flash is for, what to avoid giving it, and the facts behind that.
- What to check before trusting it
- Standard-tier input and output pricing is scheduled to increase on 1 January 2027; re-check the live rate before treating today's price as durable.
A serious default for coding-heavy Gemini work today. Test whether Google’s reported coding and reasoning improvements show up on your own task, and confirm both models’ current billing before comparing costs.
Use it for
- Long-horizon software engineering and multi-step coding loops
- Multi-step reasoning in specialised domains such as finance and legal work
- Fast multimodal document and code work
Do not treat the current rate as permanent: Google's own pricing page schedules a step-up on 1 January 2027, so a workflow costed today should be re-checked before that date.
Effort without the jargon
You need a broadly capable, fast Gemini model and will validate the output in its final workflow.
Advanced recordCost, limits, privacy and operations
Current API rates are scheduled to rise on 1 January 2027. Measure the full request cost, including thinking and tools, before budgeting.
- API cost
- Officially disclosedPaid standard API rate: US$0.75 per million input tokens and US$3.75 per million output tokens including thinking, through 31 December 2026, rising to US$1.50 and US$7.50 from 1 January 2027.
- Context window
- Officially disclosed1,048,576 input tokens
- Maximum output
- Officially disclosed65,536 output tokens
- Knowledge cutoff
- Not disclosedNot stated on the model card
- Recorded inputs
- Text · Images · Video · Audio · PDF
- Recorded output
- Text
Data handling is a product decision
Depends on surfaceData-use conditions differ by free and paid Gemini API tiers and by Google product surface; verify the live pricing and service terms for the route you use.
Operational constraints
- Standard-tier input and output pricing is scheduled to increase on 1 January 2027; re-check the live rate before treating today's price as durable.
- Gemini 3.8 Flash Cyber is a separate, access-restricted variant through the Fairwind Program, not something available on the standard API.
- The Gemini app, AI Mode in Search, AI Studio, the Gemini API and Gemini Enterprise are different operational surfaces.
Record details
- Evidence readiness
- informational
- Guidance confidence
- moderate
- Evidence revision
- gemini-38-flash@v2
Unknown means the cited official record does not disclose a safe value. It is not an estimate. Prices are provider list prices in USD where stated and can change before this record's review date.
Keep the evidence separate
Google calls Gemini 3.8 Flash its most intelligent workhorse model yet for coding and agents, reporting significant gains over 3.7 Flash on long-horizon software engineering benchmarks and on multi-step reasoning in specialised domains. It also introduced Gemini 3.8 Flash Cyber, a cybersecurity-specialised variant restricted to its Fairwind Program for vetted government, critical-infrastructure and software-maintainer defenders.
No sufficiently useful independent note has been added yet. That absence is not filled with our guess.
No Human Bit result is claimed; use this as evaluation guidance only.
Not scheduled
No Human Bit result is claimed for this model yet.
Reviewed comparisons that include Gemini 3.8 Flash
A practical comparison for ordinary office work, covering documents and email, research with sources, finished files, voice, free plans, and the question of whose account your words end up in.
Product comparisonClaude vs GeminiA practical comparison for work that involves long documents and writing that has to land, set against the convenience of an assistant that already sits inside Gmail, Drive and Calendar.