GoogleGoogle·text

Gemini 3.1 Pro Preview

AudioVisionReasoningWeb searchFunction calling
Advertised as gemini-3-1-pro-previewgemini-3.1-pro-preview
Quick reference
Gemini 3.1 Pro Preview — TLDR
  • 🧠 Google's most advanced reasoning model, built on Gemini 3 Pro.
  • 📏 1M-token context spanning text, images, audio, video, code.
  • 👁️ Multimodal input: text, image, speech, and video.
  • 🔧 Dedicated custom-tools endpoint prioritizes bash and custom functions.
  • 📚 Handles PDFs and entire code repositories in context.
  • 🏢 Available through Google Cloud's Gemini Enterprise Agent Platform.
  • ⚡ Faster than average and fairly concise, per Artificial Analysis.
  • 📅 Preview release February 2026, ahead of general availability.
💰 Best price on AntSeed
$0.227 / $1.3689%
per 1M · cheapest in / out
📏 Context
1M tokens
🐜 Sellers
13
advertising on AntSeed
Provider

Google is an American multinational technology corporation and one of the world's most valuable brands. A subsidiary of parent company Alphabet Inc., Google operates across search, cloud computing, consumer electronics, and artificial intelligence. Its DeepMind and Google…

Explore 17 more models by Google
About this model

Gemini 3.1 Pro Preview is the latest iteration of Google's flagship "Pro" frontier line, positioned by Google DeepMind as its most advanced reasoning model for complex, multi-step problems. It comprehends massively multimodal inputs — text, audio, images, video, PDFs, and entire code repositories — within a 1M-token context window, outputting up to roughly 66K tokens. Google states the model is based on its predecessor, Gemini 3 Pro, which launched in November 2025.

According to Google Cloud's documentation, 3.1 Pro exposes a dedicated custom-tools endpoint that prioritizes bash and custom functions such as file viewing and code search, aimed at agentic workflows. It is offered in preview via the Gemini Enterprise Agent Platform, allowing teams to validate the updates before general availability.

Within the broader family, 3.1 Pro is the reasoning-heavy counterpart to the faster [[sibling:gemini-3-5-flash|Gemini 3.5 Flash]] and earlier [[sibling:gemini-3-flash-preview|Gemini 3 Flash Preview]], which trade some depth for speed. Independent evaluator Artificial Analysis notes the model accepts text, image, speech, and video input, is faster than average and fairly concise, though somewhat expensive relative to similarly priced peers.

Sources
docs.cloud.google.comGemini 3.1 Pro | Gemini Enterprise Agent Platform | Google Cloud Documentation· docs.cloud.google.cominworld.aiGemini 3.1 PRO Preview by Google· inworld.aideepmind.googleGemini 3.1 Pro - Model Card — Google DeepMind· deepmind.googleartificialanalysis.aiGemini 3.1 Pro Preview - Intelligence, Performance & Price Analysis· artificialanalysis.ai

This About section is AI-generated from public sources via VeniceStats + Venice inference, with no human editing. It may contain inaccuracies.

Usage on AntSeed
Tokens served
2.15M
input + output
Requests
550
settled calls
Buyers
10
distinct, on this model
Sellers used
8
of 13 advertising
Settled
$2.80
gross USDC, this model
Sellers serving Gemini 3.1 Pro Preview (13)compare on the network explorer →
SellerReputationRoutingInput $/MCached $/MOutput $/MCategoriesAPI

"Best price" and the seller table are live AntSeed catalog data (advertised $/1M tokens — or $ per generated image for unit-billed image models — not settled amounts). Reputation = on-chain trust (0-100). "Routing" = the SDK's default buyer routing (what the VPR desktop app ships with): a trust ≥ 60 gate on the effective reputation, then cheapest-first among routable sellers; live failover state (per-peer cooldowns) is buyer-side runtime and not included. Model knowledge (TLDR, provider, About) via the VeniceStats enrichment layer. Advertised catalog, not the model used in any specific purchase. "Usage on AntSeed" counts only settlements whose buyers share the per-model split on-chain (metadata v2/v3, opt-in), so every usage figure is a lower bound.