Google (Gemini) Gemini 3.1 Pro Preview — what it is, when to use it and what to use instead

Summary

Gemini 3.1 Pro Preview is the preview release of Google's Gemini 3 Pro series, at $2 per million input tokens and $12 per million output tokens. It takes text, image, video, audio and PDF input within a 1,048,576-token context window and is aimed at software engineering and tool-heavy agentic work.

Pro is Google's reasoning-focused series, above Flash. Gemini 3.1 Pro Preview is listed with a February 2026 update and is named on Google's deprecation page as the replacement for the retired Gemini 2.5 Pro preview snapshots. A separate endpoint variant, gemini-3.1-pro-preview-customtools, is tuned for agentic workflows that use custom tools and bash.

It supports structured outputs, caching, function calling, code execution, search grounding, URL context, thinking and the Batch API, with a 65,536-token output limit. The Live API is not supported and it does not generate images or audio.

It is a preview model. Google's deprecation history shows preview models being shut down on shorter timelines than stable releases, so treat the model ID as provisional and keep the migration path open.

Modality

• Text
• Vision

Use case

• Chat
• Coding
• Reasoning
• Agents & tool use
• Long context

Tier

• Frontier

Access

• Hosted API

Specifications

Input price

$2.00 / M tokens

Output price

$12.00 / M tokens

Price band

Premium

Context window

1M tokens

Max output

65.5K tokens

Input

Text, image

Output

Text

Knowledge cutoff

unknown

Tool use

Yes

Structured output

Yes

Weights licence

not published

Model maker

Google (Gemini)

“unknown” means the figure has not been published, or we have not verified it yet — never that the answer is no. “not published” under the weights licence means this model is not offered with downloadable weights. The price band is ours: it bands the blended price (three parts input to one part output) as budget (≤ $0.50), standard (≤ $3), premium (≤ $12) or flagship.

Lifecycle

Google (Gemini) has not announced a deprecation or shutdown date for this model. That is a statement about what has been published, not a guarantee — providers typically announce a shutdown months ahead, and this page updates when they do.

1. Released 19 February 2026
2. Deprecated not announced
3. Retired not announced

Use it for

• Run software engineering tasks that need careful tool orchestration across many steps.
• Evaluate the Gemini 3 Pro series before a stable release lands.
• Migrate off the retired Gemini 2.5 Pro preview snapshots, for which Google names this as the replacement.

Avoid it when

• You need a stable model ID for a long-lived production integration; this is a preview release.
• You need more than 65,536 output tokens in a single response.
• Cost per call is a constraint; Gemini 3.8 Flash costs $0.75 / $3.75 against $2 / $12 here.

Alternatives to consider

• Gemini 3.8 Flash Curated

  Google (Gemini) ● Active

  A stable Gemini 3 model at $0.75 / $3.75 with the same limits and modalities, and no preview risk.

  $0.75 in · $3.75 out / M tokens · 1M context

• Gemini 2.5 Pro Curated

  Google (Gemini) ● Active

  The stable Pro release at $1.25 / $10, cheaper but with a January 2025 knowledge cutoff.

  $1.25 in · $10.00 out / M tokens · 1M context

• Claude Opus 5 Curated

  Anthropic ● Active

  Anthropic's top general tier at $5 / $25 with a 1M-token window and a stable model ID.

  $5.00 in · $25.00 out / M tokens · 1M context

• GPT-5.6 Sol Curated

  OpenAI ● Active

  OpenAI's flagship tier at $4 / $20 with a 1.05M-token window and a February 2026 knowledge cutoff.

  $4.00 in · $20.00 out / M tokens · 1.1M context

• Claude Opus 4.5 Derived

  Amazon Bedrock ● Active

  Alternative from Amazon Bedrock, same tier (frontier), similar price; shares chat, coding, reasoning, agents and long-context.

  $5.00 in · $25.00 out / M tokens · 200K context

Curated suggestions were picked by a reviewer, with the reason written by hand. Derived suggestions are ranked from the data — shared tags, the same tier and a comparable price band — not from benchmark scores. Check the capability fit yourself: we track lifecycle, price and specifications.

Sources

• ai.google.dev ↗ https://ai.google.dev/gemini-api/docs/models
• ai.google.dev ↗ https://ai.google.dev/gemini-api/docs/pricing
• ai.google.dev ↗ https://ai.google.dev/gemini-api/docs/deprecations
• ai.google.dev ↗ https://ai.google.dev/gemini-api/docs/models/gemini-3.1-pro-preview

Verified against the source on 10 September 2026. Dataset snapshot 10 September 2026. Providers can change a date without notice — the link above is always the final word.

About this profile

The specifications and dates above are facts from Google (Gemini), with the sources listed. The summary, the use-for and avoid-when lists and the curated alternatives are our assessment: editorial profile reviewed on 21 September 2026 by llm-researcher.