Gemini 3.8 Flash
API ID gemini-3.8-flash
gemini-3.8-flash is a Google model, released September 2026. It has a 1,048,576-token context window and up to 65,536 output tokens. It takes image and PDF input and supports function calling, structured outputs and reasoning. Standard price: $0.75 input and $3.75 output per 1M tokens.
Gemini 3.8 Flash
- Status
- Active
- Released
- September 2026
- Shutdown
- none announced
- Replacement
- —
No deprecation or shutdown date has been announced as of October 15, 2026.
Free tier: in place when tracking began (October 7, 2026) (source) · All free models
Benchmarks
Scores as each source published themComparing with Gemini 3.7 Flash, the previous model in this line. A score is compared only with the same benchmark from the same source, at the same setting where both have it. Full comparison: prices, limits, features
Arena
| Arena | Gemini 3.8 Flash | Gemini 3.7 Flash | Change | Rank | Date | Source |
|---|---|---|---|---|---|---|
TextBy category
| 1496 ±5 | 1487 ±5 | +9 within range | 9 of 415 35,670 votes | Oct 13, 2026 | source |
| Text: Coding | 1527 ±7 | 1516 ±8 | +11 within range | 25 of 410 9,370 votes | Oct 13, 2026 | source |
| WebDev | 1585 ±8 | 1590 ±7 | −5 within range | 32 of 142 10,077 votes | Oct 13, 2026 | source |
| Vision | 1289 ±11 | 1297 ±10 | −8 within range | 14 of 165 3,792 votes | Oct 9, 2026 | source |
| Agent | 0.040 ±0.011 | -0.004 ±0.009 | +0.044 | 14 of 54 37,462 sessions | Oct 14, 2026 | source |
Rated as gemini-3.8-flash-high (high effort); Gemini 3.7 Flash as gemini-3.7-flash-high (high effort). Text, WebDev and Vision ratings come from people voting between two anonymous answers; Text and Vision are style-controlled, as on arena.ai. Agent is Arena's separate score for real agent sessions, on its own scale. Data: Arena (arena.ai), CC BY 4.0.
Epoch AI
| Benchmark | Gemini 3.8 Flash | Gemini 3.7 Flash | Change | Setting | Run | Source |
|---|---|---|---|---|---|---|
| FrontierMath Tier 4 | 22.0% | 36.6% | −14.6 pts | high effort | Sep 2, 2026 | source |
| FrontierMath Tiers 1-3 | 68.4% | 71.6% | −3.2 pts | high effort | Sep 2, 2026 | source |
| GPQA Diamond | 95.4% | 94.8% | +0.6 pts | high effort | Sep 2, 2026 | source |
| OTIS Mock AIME 2024-2025 | 98.9% | 97.2% | +1.7 pts | high effort | Sep 2, 2026 | source |
| SimpleQA Verified | 69.7% | 69.2% | +0.5 pts | high effort | Sep 2, 2026 | source |
Run independently by Epoch AI. Data: Epoch AI, Capabilities & benchmarking (epoch.ai/benchmarks), CC BY.
On cloud platforms
All platform dates| Platform | Platform ID | State | Shutdown | Time left | Replacement | Source |
|---|---|---|---|---|---|---|
| Google Cloud Vertex AI | gemini-3.8-flash | Active | not before Sep 2, 2027 | — | source |
Read daily from each platform's own pages. The shutdown date and countdown on this page use Google's own date.
Features
Compare models| Feature | Value | Since | Source |
|---|---|---|---|
| Structured outputsThe model can be held to a JSON schema you supply, so every reply parses without retries or cleanup. | Supported | in place when tracking began (October 6, 2026) | source |
| Function callingAlso called tool use. You describe functions; the model replies with a function name and arguments for your code to run, then uses the result. | Supported | in place when tracking began (October 6, 2026) | source |
| Prompt cachingA prompt prefix you send again and again (system prompt, documents, tools) is cached, so later requests cost less and start faster. | Supported | in place when tracking began (October 6, 2026) | source |
| Batch APISend many requests as one job that runs in the background, usually finished within 24 hours, at a lower price than live requests. | Supported | in place when tracking began (October 6, 2026) | source |
| Image inputImages such as screenshots, photos and charts can go in the prompt for the model to read. | Supported | in place when tracking began (October 6, 2026) | source |
| PDF inputPDF files can go in the prompt directly, without converting them to text first. | Supported | in place when tracking began (October 6, 2026) | source |
| Reasoning / thinkingThe model can work through a problem step by step before it answers: usually better answers to hard questions, at the cost of more tokens and time. | Supported (minimal is not supported and returns an error) | in place when tracking began (October 6, 2026) | source |
| Built-in web searchThe company's API can search the web for the model during a request and cite its sources; you don't need a search service of your own. | Supported | in place when tracking began (October 6, 2026) | source |
| Context windowThe most the model can work with in one request, in tokens: your prompt, conversation history and files. A token is about three-quarters of an English word. | 1,048,576 tokens | in place when tracking began (October 6, 2026) | source |
| Max output tokensThe longest reply the model can write in one response, in tokens. A token is about three-quarters of an English word. | 65,536 tokens | in place when tracking began (October 6, 2026) | source |
Not published by Google: Fine-tuning, temperature parameter, top_p parameter, logprobs.
Price history
USD per 1M tokens, standard tierWe started recording this model's prices on Oct 1, 2026. The history grows with each daily check.
Show as a table
| From | Type | Price |
|---|---|---|
| Oct 1, 2026 | Input | $0.75 |
| Oct 1, 2026 | Output | $3.75 |
Prices
USD per 1M tokens, every type and band, by tier| Type | Price $/1M | In effect |
|---|---|---|
| Standard | ||
| input | $0.75 | until Dec 31, 2026 |
| input | $1.50 | from Jan 1, 2027 |
| cached input | $0.075 | until Dec 31, 2026 |
| cached input | $0.15 | from Jan 1, 2027 |
| output | $3.75 | until Dec 31, 2026 |
| output | $7.50 | from Jan 1, 2027 |
| cache storage per hour | $0.50 | until Dec 31, 2026 |
| cache storage per hour | $1.00 | from Jan 1, 2027 |
| Batch | ||
| input | $0.375 | until Dec 31, 2026 |
| input | $0.75 | from Jan 1, 2027 |
| cached input | $0.0375 | until Dec 31, 2026 |
| cached input | $0.075 | from Jan 1, 2027 |
| output | $1.875 | until Dec 31, 2026 |
| output | $3.75 | from Jan 1, 2027 |
| cache storage per hour | $0.50 | until Dec 31, 2026 |
| cache storage per hour | $1.00 | from Jan 1, 2027 |
| Flex | ||
| input | $0.375 | until Dec 31, 2026 |
| input | $0.75 | from Jan 1, 2027 |
| cached input | $0.0375 | until Dec 31, 2026 |
| cached input | $0.075 | from Jan 1, 2027 |
| output | $1.875 | until Dec 31, 2026 |
| output | $3.75 | from Jan 1, 2027 |
| cache storage per hour | $0.50 | until Dec 31, 2026 |
| cache storage per hour | $1.00 | from Jan 1, 2027 |
| Priority | ||
| input | $1.35 | until Dec 31, 2026 |
| input | $2.70 | from Jan 1, 2027 |
| cached input | $0.135 | until Dec 31, 2026 |
| cached input | $0.27 | from Jan 1, 2027 |
| output | $6.75 | until Dec 31, 2026 |
| output | $13.50 | from Jan 1, 2027 |
| cache storage per hour | $0.50 | until Dec 31, 2026 |
| cache storage per hour | $1.00 | from Jan 1, 2027 |
Changelog
Changes we detected on Google's pages- Platform date set
Google Cloud Vertex AI set a shutdown date for gemini-3.8-flash: not before Sep 2, 2027
Info
Related news
Email alerts
Get an email if gemini-3.8-flash changes
We'll email you the same morning we spot a deprecation notice, a new or moved shutdown date, a price change, or the model leaving Google's model list.
No password: one email to confirm, and you can unsubscribe any time. Privacy
Last checked · Sources: Google deprecations, Google pricing, Google models, Google aliases, Google api models