NOUMENOpen app
NOUMEN

Google · Frontier

Gemini 3.1 Pro (Preview)

Added Feb 19, 2026

Compare models

Choose up to 3 models to compare.0 of 3 chosen

Loading models…

Use in the app

Gemini 3.1 Pro preview is built for tasks where simple answers are not enough. Stronger core reasoning for complex coding, math, and long-context workflows, with multimodal support and a reported 77.1% verified score on ARC-AGI-2. NOTE: Inputs > 200k tokens are charged at 2x input and 1.5x output rates.

  • Reasoning
  • Tools
  • Vision
  • Structured output
  • Audio input
  • Video input
Input
Text, Images, Video, Audio
Output
Text
Reasoning effort
low · medium · high
Discovery tier
Frontier

Pricing

USD per 1M tokens
Input
$2.00
Output
$12.00
Cache read
$0.20
Cache write
$0.38

Reference prices as published by the model provider. API example

Specifications

Context window
1M
Max output
65.5K
Average output (7 days)
699 tokens
Longer than 62% of 468 models

Side by side

Compare models

Put Gemini 3.1 Pro (Preview) next to up to three other models. The comparison stays in the link, so you can share it.

Compare models

Choose up to 3 models to compare.0 of 3 chosen

Loading models…

TryGPT 6 LunaClaude Sonnet 4.6 ThinkingMiniMax M3All 3

Public evaluations

Benchmarks

Independent evaluation results, with how each score ranks among the models in the Noumen catalog.

Intelligence Index

29.7

Better than 71% of 326 scored models

3.8Catalog models by score55.1

Indexes

  • Coding Index

    68.8

    Composite of the coding evaluations

    Better than 66% of 168 scored models

Agentic work

  • AutomationBench-AA

    35.4%

    Workflow automation, penalized for guardrail violations

    Better than 52% of 173 scored models

  • AutomationBench-AA tasks completed

    9.4%

    Workflows finished end to end without a guardrail violation

    Better than 50% of 173 scored models

  • Harvey LAB-AAv1.1

    0.0%

    Legal tasks passed in full, with no material hallucination

    Better than 0% of 29 scored models

  • AA-Briefcasev1.1

    456 Elo

    Agentic knowledge work, rated head to head

    Better than 15% of 166 scored models

  • GDPval-AAv2.1

    791 Elo

    Economically valuable occupational tasks, rated head to head

    Better than 25% of 187 scored models

Document reasoning

  • GDP.pdf

    17.8%

    Professional reasoning over PDF documents, all-pass rate

    Better than 70% of 171 scored models

  • AA-LCRv1.1

    82.0%

    Reasoning across long documents

    Better than 85% of 309 scored models

  • MLCR-AA

    11.1%

    Long-context reasoning over medical records

    Better than 37% of 124 scored models

Reasoning and knowledge

  • Humanity’s Last Exam

    47.0%

    Expert-level questions across academic fields

    Better than 90% of 323 scored models

  • GPQA Diamond

    94.1%

    Graduate-level science questions

    Better than 96% of 291 scored models

  • IFBench

    77.1%

    Following precise output instructions

    Better than 93% of 215 scored models

Coding

  • Terminal-Benchv4.0

    4.0%

    Practical software engineering in a terminal

    Better than 51% of 171 scored models

  • SciCode

    58.7%

    Scientific computing in Python

    Better than 85% of 174 scored models

Legacy evaluations

  • Terminal-Bench Hard

    53.8%

    Earlier terminal task suite

    Better than 93% of 212 scored models

  • τ²-Bench Telecom

    95.6%

    Tool use in a customer-support simulation

    Better than 87% of 214 scored models

  • AA-LCR (unversioned)

    82.0%

    Long-context reasoning, earlier grading

    Better than 85% of 309 scored models

Scores from Artificial Analysis, updated Oct 26, 2026. Artificial Analysis

Live routing data

Performance and reliability

Recent averages across requests routed for this model. Your own speed depends on prompt length, reasoning effort and load.

Throughput
96.9 tok/s
Faster than 60% of 263 models
Time to first token
5.7 s
Quicker than 24% of 263 models
Prompt cache hits
28.8%
About $1.48 per 1M input tokens with caching
Request success (24h)
99.8%
Reliable

Cache hits are the share of input tokens recently served from a prompt cache across all requests. They are not a forecast of your own hit rate.

Hourly request success · last 72 hoursLowest hour 97.2%
  • Healthy (98% or more)
  • Degraded (90–98%)
  • Down (under 90%)
  • No requests

Developers

Use it through the API

Noumen’s API is OpenAI-compatible: any Chat Completions client works with a base URL and a key.

Read the quickstart
curl https://www.noumen.xyz/api/v1/chat/completions \
  -H "Authorization: Bearer $NOUMEN_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "google/gemini-3.1-pro-preview",
    "reasoning_effort": "medium",
    "messages": [{ "role": "user", "content": "Explain this stack trace." }]
  }'
import os
from openai import OpenAI

client = OpenAI(base_url="https://www.noumen.xyz/api/v1", api_key=os.environ["NOUMEN_API_KEY"])
reply = client.chat.completions.create(
    model="google/gemini-3.1-pro-preview",
    messages=[{"role": "user", "content": "Explain this stack trace."}],
)
print(reply.choices[0].message.content)

Google

Other Google models, starting with the same family.