Home/Models/Google/gemini-3.1-pro
G

gemini-3.1-pro

Input:$1.6/M
Output:$9.6/M
Gemini 3.1 Pro is the next generation in the Gemini series of models, a suite of highly-capable, natively multimodal, reasoning models. Gemini 3 Pro is now Google’s most advanced model for complex tasks, and can comprehend vast datasets, challenging problems from different information sources, including text, audio, images, video, and entire code repositories
New
Commercial Use
Playground
Overview
Features
Pricing
API
Versions

Technical specifications — Gemini 3.1 Pro

Itemgemini-3-pro (public summary)
ProviderGoogle
Canonical model idgemini-3-pro (public preview)
Input typesText, Image, Video, Audio, PDF
Output typesText (natural language, structured outputs, function-call payloads)
Input token limit (context)1,048,576 tokens
Output token limit65,536 tokens
Function-calling / tool useSupported (function calling, structured outputs, tool integrations)
MultimodalityFull multimodal support (images, video, audio, documents)
Code execution & agentic flowsSupported (agent mode, code assist, tool orchestration)
Knowledge cutoffJanuary 2025

What is Gemini 3.1 Pro ?

Gemini 3.1 Pro is Google’s publicly flagship in the Gemini 3 family, positioned as a state-of-the-art multimodal reasoning model with advanced agentic and developer tooling. The model emphasizes high-capacity context handling (over 1M token inputs), broad media support (images, video, audio, PDF), and deep integrations for tool use, function calling, and code-centric workflows (e.g., Gemini Code Assist and agent modes).

Gemini 3 Pro is presented by Google as optimized for both interactive developer experiences (low-latency coding and agent workflows) and high-fidelity multimodal understanding (interpreting and reasoning across mixed media inputs).

Main Features of Gemini 3.1 Pro

Gemini-3.1 Pro (via its Preview) introduces the following features:

Multimodal Integration

Processes inputs across:

  • Natural language
  • Images
  • Speech/audio
  • Video

with a unified token representation for cross-modal reasoning.

Extended Context Window

An exceptionally large context capacity of up to ~1 million tokens enables handling of:

  • Long documents
  • Multidocument synthesis
  • Codebases and transcripts.

This surpasses many competing models that typically support ~32 K–262 K tokens.

Sparse Mixture-of-Experts (MoE) Scaling

Sparse MoE routing allows scaling internal model capacity without proportional compute costs, improving reasoning at scale.

Advanced Reasoning / Planning

Innovations like chain-of-thought training, reinforcement learning from human feedback, and specialized benchmarks make it strong on logical and mathematical tasks.

Supposed benchmarks:

AIME 2025: 100% (with code execution)
SWE-Bench Verified: 83.9%
ARC-AGI-2: 71.8%
LiveCodeBench Pro: 2844 Elo
Terminal-Bench 2.0: 63.5%
MMMLU: 93.6%

Representative enterprise use cases

  • End-to-end media pipelines: Ingest video, transcript, and images to produce synchronized summaries, metadata, and structured insight at scale.
  • Large-scale code generation and review: Use in IDEs and CI pipelines to auto-generate code, refactor multi-file projects, and produce test suggestions across large codebases.
  • Agentic automation: Coordinate multi-tool agents that interact with cloud services, orchestration systems, and internal APIs using structured function calls.
  • Research & content production: Draft long-form content (reports, books) that combine text and embedded multimedia with internal cross-references preserved.

How to access Gemini 3.1 Pro API

Step 1: Sign Up for API Key

Log in to cometapi.com. If you are not our user yet, please register first. Sign into your CometAPI console. Get the access credential API key of the interface. Click “Add Token” at the API token in the personal center, get the token key: sk-xxxxx and submit.

Step 2: Send Requests to Gemini 3.1 Pro API

Select the “gemini-3.1-pro” endpoint to send the API request and set the request body. The request method and request body are obtained from our website API doc. Our website also provides Apifox test for your convenience. Replace <YOUR_API_KEY> with your actual CometAPI key from your account. base url is Gemini Generating Content and Chat.

Insert your question or request into the content field—this is what the model will respond to . Process the API response to get the generated answer.

Step 3: Retrieve and Verify Results

Process the API response to get the generated answer. After processing, the API responds with the task status and output data.

See also Gemini 3 Pro API

FAQ

Can the Gemini 3.1 Pro API handle 1,048,576‑token documents?

Yes. The Gemini 3.1 Pro supports a context window of up to 1,048,576 tokens and can produce outputs up to 65,536 tokens, enabling single‑session reasoning over very large documents or codebases.

What input modalities does Gemini 3.1 Pro accept through the API?

Gemini 3.1 Pro Preview accepts text, images, video, audio, and PDF inputs and can reason across these modalities in a single session.

Does Gemini 3.1 Pro support function calling, structured outputs, and agentic tool use?

Yes. The API supports function calling, JSON‑constrained structured outputs, search grounding, and agentic/tool execution patterns (including code execution hooks in supported environments).

Is Gemini 3.1 Pro production‑ready?

Exercise caution: it is a public preview release (launched Nov 18, 2025). Validate API contracts, quotas, and behavior for your workloads and sandbox agentic or code‑execution features before wide production rollout.

What is Gemini 3.1 Pro Preview's knowledge cutoff date?

The documented knowledge cutoff for Gemini 3 Pro Preview is January 2025.

Features for gemini-3.1-pro

Explore the key features of gemini-3.1-pro, designed to enhance performance and usability. Discover how these capabilities can benefit your projects and improve user experience.

Pricing for gemini-3.1-pro

Explore competitive pricing for gemini-3.1-pro, designed to fit various budgets and usage needs. Our flexible plans ensure you only pay for what you use, making it easy to scale as your requirements grow. Discover how gemini-3.1-pro can enhance your projects while keeping costs manageable.
Comet Price (USD / M Tokens)Official Price (USD / M Tokens)Discount
Input:$1.6/M
Output:$9.6/M
Input:$2/M
Output:$12/M
-20%

Sample code and API for gemini-3.1-pro

Access comprehensive sample code and API resources for gemini-3.1-pro to streamline your integration process. Our detailed documentation provides step-by-step guidance, helping you leverage the full potential of gemini-3.1-pro in your projects.
Python
JavaScript
Curl
from google import genai
import os

# Get your CometAPI key from https://www.cometapi.com/console/token, and paste it here
COMETAPI_KEY = os.environ.get("COMETAPI_KEY") or "<YOUR_COMETAPI_KEY>"
BASE_URL = "https://api.cometapi.com"

client = genai.Client(
    http_options={"api_version": "v1beta", "base_url": BASE_URL},
    api_key=COMETAPI_KEY,
)

response = client.models.generate_content(
    model="gemini-3.1-pro-preview",
    contents="Explain how AI works in a few words",
)

print(response.text)

Versions of gemini-3.1-pro

The reason gemini-3.1-pro has multiple snapshots may include potential factors such as variations in output after updates requiring older snapshots for consistency, providing developers a transition period for adaptation and migration, and different snapshots corresponding to global or regional endpoints to optimize user experience. For detailed differences between versions, please refer to the official documentation.
version
gemini-3.1-pro-preview
gemini-3.1-pro-preview-thinking

More Models