NOUMENOpen app
NOUMEN

Google · Balanced

Gemini 2.5 Pro Preview 0325

Added Mar 25, 2025

Compare models

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

Loading models…

Use in the app

Gemini 2.5 Pro Preview 0325. Google's model from March 2025.

  • Reasoning
  • Vision
  • Audio input
Input
Text, Images, Audio
Output
Text
Reasoning effort
low · medium · high
Discovery tier
Balanced

Pricing

USD per 1M tokens
Input
$2.50
Output
$10.00
Cache read
$0.25

Reference prices as published by the model provider. API example

Specifications

Context window
1M
Max output
65.5K
Average output (7 days)
1.3K tokens
Longer than 81% of 465 models

Side by side

Gemini 2.5 Pro Preview 0325 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 Pro Preview 0325GoogleThis modelQwen3 MaxQwenRemove
ProviderGoogleQwen
Discovery tierBalancedBalanced
AddedMar 25, 2025Sep 5, 2025
Limits
Context window1.05MBest256K
Max output65.54KBest32.77K
Average output1.3K tokens—
Pricing, USD per 1M tokens
Input$2.50$1.20Best
Output$10.00$6.00Best
Cache read$0.25—
Benchmark indexes
Intelligence Index16.1Best15.6
Coding Index33.3—
Math Index87.7Best80.7
Capabilities
ReasoningYesNo
VisionYesNo
Audio 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 Preview 0325
  • Qwen3 Max

Intelligence Index

  • Gemini 2.5 Pro Preview 032516.1
  • Qwen3 Max15.6

Gemini 2.5 Pro Preview 0325: 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 Preview 0325: 33.3
    • Qwen3 Max: —

  • Math Index

    Composite of the math evaluations

    • Gemini 2.5 Pro Preview 0325: 87.7Top
    • Qwen3 Max: 80.7

Agentic work

  • AutomationBench-AA

    Workflow automation, penalized for guardrail violations

    • Gemini 2.5 Pro Preview 0325: 2.2%
    • Qwen3 Max: —

  • AutomationBench-AA tasks completed

    Workflows finished end to end without a guardrail violation

    • Gemini 2.5 Pro Preview 0325: 0.0%
    • Qwen3 Max: —

  • AA-Briefcasev1.1

    Agentic knowledge work, rated head to head

    • Gemini 2.5 Pro Preview 0325: 302 Elo
    • Qwen3 Max: —

  • GDPval-AAv2.1

    Economically valuable occupational tasks, rated head to head

    • Gemini 2.5 Pro Preview 0325: 455 Elo
    • Qwen3 Max: —

Document reasoning

  • GDP.pdf

    Professional reasoning over PDF documents, all-pass rate

    • Gemini 2.5 Pro Preview 0325: 10.2%
    • Qwen3 Max: —

  • AA-LCRv1.1

    Reasoning across long documents

    • Gemini 2.5 Pro Preview 0325: 69.0%Top
    • Qwen3 Max: 50.0%

  • MLCR-AA

    Long-context reasoning over medical records

    • Gemini 2.5 Pro Preview 0325: 12.2%
    • Qwen3 Max: —

Reasoning and knowledge

  • Humanity’s Last Exam

    Expert-level questions across academic fields

    • Gemini 2.5 Pro Preview 0325: 22.5%Top
    • Qwen3 Max: 11.9%

  • GPQA Diamond

    Graduate-level science questions

    • Gemini 2.5 Pro Preview 0325: 84.4%Top
    • Qwen3 Max: 76.4%

  • MMLU-Pro

    Multi-subject knowledge and reasoning

    • Gemini 2.5 Pro Preview 0325: 86.2%Top
    • Qwen3 Max: 84.1%

  • IFBench

    Following precise output instructions

    • Gemini 2.5 Pro Preview 0325: 48.7%Top
    • Qwen3 Max: 44.1%

Coding

  • Terminal-Benchv4.0

    Practical software engineering in a terminal

    • Gemini 2.5 Pro Preview 0325: 0.0%
    • Qwen3 Max: —

  • SciCode

    Scientific computing in Python

    • Gemini 2.5 Pro Preview 0325: 46.3%
    • Qwen3 Max: —

  • LiveCodeBench

    Recent competitive programming problems

    • Gemini 2.5 Pro Preview 0325: 80.1%Top
    • Qwen3 Max: 76.7%

Math

  • AIME 2025

    Competition mathematics, 2025 problems

    • Gemini 2.5 Pro Preview 0325: 87.7%Top
    • Qwen3 Max: 80.7%

  • AIME

    Competition mathematics

    • Gemini 2.5 Pro Preview 0325: 88.7%
    • Qwen3 Max: —

  • MATH-500

    Competition math problems

    • Gemini 2.5 Pro Preview 0325: 96.7%
    • Qwen3 Max: —

Legacy evaluations

  • Terminal-Bench Hard

    Earlier terminal task suite

    • Gemini 2.5 Pro Preview 0325: 26.5%Top
    • Qwen3 Max: 20.5%

  • τ²-Bench Telecom

    Tool use in a customer-support simulation

    • Gemini 2.5 Pro Preview 0325: 54.1%
    • Qwen3 Max: 74.3%Top

  • AA-LCR (unversioned)

    Long-context reasoning, earlier grading

    • Gemini 2.5 Pro Preview 0325: 69.0%Top
    • Qwen3 Max: 50.0%

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

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-preview-03-25",
    "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-preview-03-25",
    messages=[{"role": "user", "content": "Explain this stack trace."}],
)
print(reply.choices[0].message.content)

Google

Other Google models, starting with the same family.