Gemini 3.1 Pro
A preview model for complex reasoning, coding, multimodal understanding, and very large inputs.
Gemini Pro · Preview
Overview
Gemini 3.1 Pro Preview, available in Aether as google/gemini-3.1-pro-preview, is designed to improve reasoning quality, token efficiency, and reliability in software engineering and multi-step agent workflows.
The model accepts text, images, video, audio, and PDFs, making it particularly useful when an Aether conversation includes more than plain text.
+Pros
- Very large context window
- Text, image, video, audio, and PDF input
- Strong fit for software engineering and multi-step work
-Cons
- Preview status means behavior and availability can change
- Simple prompts usually do not need its full context and reasoning depth
Best suited to
Complex reasoning and planning
Large document analysis
Multimodal questions
Coding and tool-oriented work
Aether Workspace Position
Available in Aether as a preview model.
Aether can select Gemini 3.1 Pro for complex, multimodal, document-heavy, or long-context requests.
Research sources
- Gemini 3.1 Pro PreviewGoogle AI for Developers
- Gemini deprecationsGoogle AI for Developers
