Gemini 3.1 Pro Preview by Vercel AI Gateway

$10 off the fastest Kimi K3 and GLM 5.3$10 off the fastest Kimi K3 and GLM 5.3

Model details

Gemini 3.1 Pro Preview

Gemini 3.1 Pro Preview is Google's frontier reasoning model designed for complex, multi-step problem solving where accuracy takes priority over raw speed. Published evaluations highlight an "upgraded core intelligence" that produces higher scores on ARC-AGI-2, GPQA Diamond, and SWE-Bench Verified, positioning it as a notable step in the Gemini Pro lineage for tasks requiring deliberate analytical chains. Independent benchmarking placed the model at the top of the Artificial Analysis Intelligence Index while costing less than half of competing frontier offerings, underscoring its focus on balancing sophisticated reasoning with accessible deployment economics.

In practical use, Gemini 3.1 Pro Preview combines multimodal inputs with a large context window and substantial output capacity, making it well suited to document-heavy workflows, code analysis, and structured multi-turn agent loops that call external tools. The model supports function calling, web search grounding, prompt caching, and structured response schemas, giving developers a flexible foundation for building reliable assistants and automated pipelines. Hands-on coverage described the release as an immediately available frontier upgrade, and third-party routing platforms quickly expanded catalogue support, signaling broad ecosystem readiness for production experimentation.

Vercel AI Gatewaygoogle/gemini-3.1-pro-previewgemini-pro

Open provider page Provider docs

Quick Info

Provider: Vercel AI Gateway
Model key: google/gemini-3.1-pro-preview
Release date: Feb 19, 2026
Last updated: Feb 19, 2026
Knowledge cutoff: 2025-01
Input modalities
Output modalities
Capabilities

Cost

Input token cost: $2.00
Output token cost: $12.00

Limits

Output tokens: 64,000 tokens
Context window: 1,000,000 tokens

Transparent token rates

Compare gemini-pro pricing

Rates are shown per one million tokens. Combined means one million input plus one million output tokens.

Browse this family

Latest news about Gemini 3.1 Pro Preview

Vercel AI Gateway

CoveragePreview

Hong Kong-based API platform adds Google's latest multimodal model to its growing roster, expanding developer access to frontier AI through a unified...

Vercel AI Gateway

CoveragePreview

Published evals of the “upgraded core intelligence” for Gemini 3.1 Pro show higher scores on ARC-AGI-2, GPQA Diamond and SWE-Bench Verified

Vercel AI Gateway

CoveragePreview

Google's Gemini 3.1 Pro leads the Artificial Analysis Intelligence Index and costs less than half the price of the competition. But benchmarks aren't everything.

Vercel AI Gateway

Coverage

Google today announced Gemini 3.1 Pro, the latest version of its frontier AI LLM (Large Language Model). The new AI assistant is immediately available. If you&#

Vercel AI Gateway

CoverageRelease Notes

Gemini 3.1 Pro: A smarter model for your most complex tasks Google has just announced the release of Gemini 3.1 Pro, a major upgrade to its core AI in...

Vercel AI Gateway

Official sourcePreview

Vercel's official FAQ documents that Gemini 3.1 Pro Preview on the AI Gateway advances over Gemini 3 Pro Preview in three areas: software engineering and agentic workflow quality, finance and spreadsheet usability, and more efficient thinking that reduces token consumption without degrading output quality. The page con For developer integration, Vercel states that AI Gateway manages all Google provider credentials, so users only need a Vercel API key or OIDC token rather than a separate Google API account. The model is exposed through the AI Gateway API, the AI SDK (TypeScript and Python, supporting Chat Completions, Messages, and Re