Gemini 3.1 Pro Preview

GoogleMCP Tool CallReasoning

google/gemini-3.1-pro-previewClosed API

Reasoning-first Gemini preview for agentic coding and complex problem solving

Benchmark Pricing$2.00 / 1M in$12.00 / 1M out$0.20 cache read

Context: 1,048,576 tokens•Max Output: 65,536 tokens•Knowledge: 2025-01•Released: 2026-02-19

TECHNICAL SPECIFICATIONS

Context Window1,048,5761M tokens

Max Output65,53665.5K tokens

LicenseProprietaryCloud API only

Input Modalities

TextVisionPDFAudioVideo

Function / Tool CallingSupported (MCP Ready)

Structured OutputsGuaranteed Schema

MODEL CONTEXT PROTOCOL (MCP) READINESS

✓ Gemini 3.1 Pro Preview natively supports OpenAI / Anthropic format tool calling and is fully compatible with MCP servers and agent loops.

MCP Context Window Recommendations

With a context capacity of 1,048,576 tokens, this model can handle extremely large MCP resource manifests, complete database schemas, and multiple active server tool definitions simultaneously without degradation.

STANDARDIZED BENCHMARKS & EVALS

SWE-Bench Proresolve rate

54.2

Terminal-Benchsuccess rate

70.3

SWE-Bench Proresolve rate

46.1

SWE-Atlas Codebase QnAscore

13.5

SWE-Atlas Refactoringscore

33.81

SWE-Atlas Test Writingscore

29.84

Artificial Analysis Coding Agent Indexaverage pass@1

SWE-Atlas Codebase QnApass@1

45.6

SWE-Bench Propass@1

15.1

Terminal-Benchpass@1

68.3

GPQA Diamondaccuracy

94.3

Humanity's Last Examaccuracy

44.4

ARC-AGI-2accuracy

77.1

MMMU Proaccuracy

80.5

MCP Atlassuccess rate

78.2

OSWorld-Verifiedsuccess rate

76.2

CharXiv Reasoningaccuracy

83.3

GDPval-AAElo

1314

API CALL INVOCATION EXAMPLE

// Example: Calling Gemini 3.1 Pro Preview via OpenAI SDK
import OpenAI from "openai";

const client = new OpenAI({
  apiKey: process.env.GOOGLE_API_KEY,
});

const response = await client.chat.completions.create({
  model: "google/gemini-3.1-pro-preview",
  messages: [{ role: "user", content: "Hello! Can you inspect MCP tools?" }],
});

console.log(response.choices[0].message.content);