GoogleAIWS category: Fast MultimodalGeneral availability
Gemini 3.8 Flash
Google Gemini 3.8 Flash, a generally available multimodal Gemini model for long-horizon software engineering, autonomous agents, and complex knowledge workflows.
Facts Verified
Sep 2, 2026
Verification
Verified
Profile
Source-backed
Context
1,048,576 tokens
Max output
65,536 tokens
API
Available
Specifications
Context
1,048,576 tokens
Max output
65,536 tokens
CompanyGoogle
Release DateSep 2, 2026
Provider statusGeneral availability
LicenseNot published
API availabilityAvailable
Capability Map
Verified fieldsReasoning
Supported
Image input
Supported
Audio input
Supported
Video input
Supported
API access
Available
Native input modalities
textimagevideoaudiopdf
Native output: text
Built-in tools
Image generation
Not supported
Web search
Supported
File search
Supported
Code interpreter
Supported
Computer use
Supported
Native output is what the model returns directly. Tool capabilities are separate.
Developer API Rates
Verified public rateInput / 1M
$0.7500
Output / 1M
$3.75
Cached $0.0750 / 1MRate through Dec 31, 2026
Introductory paid-tier rates apply through December 31, 2026. Starting January 1, 2027, standard input/output pricing rises to $1.50/$7.50 per 1M tokens.
Key Strengths
- Long-horizon software engineering
- Agentic and tool-using workflows
- Multimodal inputs
- Low introductory token pricing
Limitations & Trade-offs
- Higher effort levels can use more tokens
- Introductory pricing expires after December 31, 2026
Provider API model ID
Use this identifier in the provider API where supported.
gemini-3.8-flash