NOUMENOpen app
NOUMEN

Google · Frontier

Gemini 2.5 Pro

Added Jun 5, 2025

Compare models

Choose up to 3 models to compare.1 of 3 chosen

Loading models…

Use in the app

Gemini 2.5 Pro stable release. Google's most capable generalist model with strong performance across a wide range of tasks.

  • Reasoning
  • Tools
  • Vision
  • 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
$1.25
Output
$10.00
Cache read
$0.13
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)
1.1K tokens
Longer than 72% of 465 models

Side by side

Gemini 2.5 Pro compared

Specifications, prices, speed and benchmark scores next to each other. Benchmarks below show every compared model too.

Compare models

Choose up to 3 models to compare.1 of 3 chosen

Loading models…

Gemini 2.5 ProGoogleThis modelDeepSeek V3.2DeepSeekRemove
ProviderGoogleDeepSeek
Discovery tierFrontierOpen weights
AddedJun 5, 2025Dec 1, 2025
Limits
Context window1.05MBest163K
Max output65.54K65.54K
Average output1.1K tokens408 tokens
Pricing, USD per 1M tokens
Input$1.25$0.21Best
Output$10.00$0.31Best
Cache read$0.13$0.0216Best
Cache write$0.38—
Performance
Throughput189.7 tok/sBest46.8 tok/s
Time to first token12.3 s3.2 sBest
Request success (24h)99.5%99.6%Best
Benchmark indexes
Intelligence Index16.1Best16.0
Coding Index33.3—
Math Index87.7Best59.0
Capabilities
ReasoningYesNo
ToolsYesYes
VisionYesNo
Structured outputNoYes
PDF inputNoYes
Audio inputYesNo
Video inputYesNo
Reasoning effortlow, medium, high—

Public evaluations

Benchmarks

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

  • Gemini 2.5 Pro
  • DeepSeek V3.2

Intelligence Index

  • Gemini 2.5 Pro16.1
  • DeepSeek V3.216.0

Gemini 2.5 Pro: Better than 41% of 326 scored models

3.8Catalog models by score55.1

Indexes

  • Coding Index

    Composite of the coding evaluations

    • Gemini 2.5 Pro: 33.3
    • DeepSeek V3.2: —

  • Math Index

    Composite of the math evaluations

    • Gemini 2.5 Pro: 87.7Top
    • DeepSeek V3.2: 59.0

Agentic work

  • AutomationBench-AA

    Workflow automation, penalized for guardrail violations

    • Gemini 2.5 Pro: 2.2%
    • DeepSeek V3.2: —

  • AutomationBench-AA tasks completed

    Workflows finished end to end without a guardrail violation

    • Gemini 2.5 Pro: 0.0%
    • DeepSeek V3.2: —

  • AA-Briefcasev1.1

    Agentic knowledge work, rated head to head

    • Gemini 2.5 Pro: 302 Elo
    • DeepSeek V3.2: —

  • GDPval-AAv2.1

    Economically valuable occupational tasks, rated head to head

    • Gemini 2.5 Pro: 455 Elo
    • DeepSeek V3.2: —

Document reasoning

  • GDP.pdf

    Professional reasoning over PDF documents, all-pass rate

    • Gemini 2.5 Pro: 10.2%
    • DeepSeek V3.2: —

  • AA-LCRv1.1

    Reasoning across long documents

    • Gemini 2.5 Pro: 69.0%Top
    • DeepSeek V3.2: 45.7%

  • MLCR-AA

    Long-context reasoning over medical records

    • Gemini 2.5 Pro: 12.2%
    • DeepSeek V3.2: —

Reasoning and knowledge

  • Humanity’s Last Exam

    Expert-level questions across academic fields

    • Gemini 2.5 Pro: 22.5%Top
    • DeepSeek V3.2: 11.2%

  • GPQA Diamond

    Graduate-level science questions

    • Gemini 2.5 Pro: 84.4%Top
    • DeepSeek V3.2: 75.1%

  • MMLU-Pro

    Multi-subject knowledge and reasoning

    • Gemini 2.5 Pro: 86.2%Top
    • DeepSeek V3.2: 83.7%

  • IFBench

    Following precise output instructions

    • Gemini 2.5 Pro: 48.7%
    • DeepSeek V3.2: 49.0%Top

Coding

  • Terminal-Benchv4.0

    Practical software engineering in a terminal

    • Gemini 2.5 Pro: 0.0%
    • DeepSeek V3.2: —

  • SciCode

    Scientific computing in Python

    • Gemini 2.5 Pro: 46.3%
    • DeepSeek V3.2: —

  • LiveCodeBench

    Recent competitive programming problems

    • Gemini 2.5 Pro: 80.1%Top
    • DeepSeek V3.2: 59.3%

Math

  • AIME 2025

    Competition mathematics, 2025 problems

    • Gemini 2.5 Pro: 87.7%Top
    • DeepSeek V3.2: 59.0%

  • AIME

    Competition mathematics

    • Gemini 2.5 Pro: 88.7%
    • DeepSeek V3.2: —

  • MATH-500

    Competition math problems

    • Gemini 2.5 Pro: 96.7%
    • DeepSeek V3.2: —

Legacy evaluations

  • Terminal-Bench Hard

    Earlier terminal task suite

    • Gemini 2.5 Pro: 26.5%
    • DeepSeek V3.2: 32.6%Top

  • τ²-Bench Telecom

    Tool use in a customer-support simulation

    • Gemini 2.5 Pro: 54.1%
    • DeepSeek V3.2: 78.9%Top

  • AA-LCR (unversioned)

    Long-context reasoning, earlier grading

    • Gemini 2.5 Pro: 69.0%Top
    • DeepSeek V3.2: 45.7%

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
189.7 tok/s
Faster than 84% of 250 models
Time to first token
12.3 s
Quicker than 7% of 250 models
Prompt cache hits
79.6%
About $0.35 per 1M input tokens with caching
Request success (24h)
99.5%
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 96.6%
  • 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": "gemini-2.5-pro",
    "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="gemini-2.5-pro",
    messages=[{"role": "user", "content": "Explain this stack trace."}],
)
print(reply.choices[0].message.content)

Google

Other Google models, starting with the same family.