NOUMENOpen app
NOUMEN

OpenAI · Frontier

GPT 5

Added Aug 7, 2025

Compare models

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

Loading models…

Use in the app

GPT-5 is OpenAI's most advanced model, offering major improvements in reasoning, code quality, and user experience. It handles complex coding tasks with minimal prompting, provides clear explanations, and introduces enhanced agentic capabilities. Designed for logic and multi-step tasks.

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

Pricing

USD per 1M tokens
Input
$1.25
Output
$10.00
Cache read
$0.13

Reference prices as published by the model provider. API example

Specifications

Context window
400K
Max output
128K
Average output (7 days)
411 tokens
Longer than 41% of 468 models

Side by side

GPT 5 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…

GPT 5OpenAIThis modelGPT LatestOpenAIRemove
ProviderOpenAIOpenAI
Discovery tierFrontierFrontier
AddedAug 7, 2025Mar 29, 2026
Limits
Context window400K1.05MBest
Max output128K128K
Average output411 tokens311 tokens
Pricing, USD per 1M tokens
Input$1.25Best$10.00
Output$10.00Best$50.00
Cache read$0.13Best$1.00
Cache write—$12.50
Performance
Throughput—234.1 tok/s
Time to first token—10.5 s
Request success (24h)—97.5%
Benchmark indexes
Intelligence Index23.049.6Best
Coding Index37.876.7Best
Math Index94.3—
Capabilities
ReasoningYesYes
ToolsYesYes
VisionYesYes
Structured outputYesYes
PDF inputYesYes
Reasoning effortminimal, low, medium, highlow, medium, high, xhigh, max

Public evaluations

Benchmarks

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

  • GPT 5
  • GPT Latest

Intelligence Index

  • GPT 523.0
  • GPT Latest49.6

GPT 5 tested at high reasoning effort

GPT 5: Better than 57% of 326 scored models

3.8Catalog models by score55.1

Indexes

  • Coding Index

    Composite of the coding evaluations

    • GPT 5: 37.8
    • GPT Latest: 76.7Top

  • Math Index

    Composite of the math evaluations

    • GPT 5: 94.3
    • GPT Latest: —

Agentic work

  • AutomationBench-AA

    Workflow automation, penalized for guardrail violations

    • GPT 5: —
    • GPT Latest: 64.6%

  • AutomationBench-AA tasks completed

    Workflows finished end to end without a guardrail violation

    • GPT 5: —
    • GPT Latest: 33.6%

  • AA-Briefcasev1.1

    Agentic knowledge work, rated head to head

    • GPT 5: —
    • GPT Latest: 1,460 Elo

  • GDPval-AAv2.1

    Economically valuable occupational tasks, rated head to head

    • GPT 5: 918 Elo
    • GPT Latest: 1,487 EloTop

Document reasoning

  • GDP.pdf

    Professional reasoning over PDF documents, all-pass rate

    • GPT 5: —
    • GPT Latest: 30.4%

  • AA-LCRv1.1

    Reasoning across long documents

    • GPT 5: 78.2%
    • GPT Latest: 79.7%Top

  • MLCR-AA

    Long-context reasoning over medical records

    • GPT 5: —
    • GPT Latest: 38.3%

Reasoning and knowledge

  • Humanity’s Last Exam

    Expert-level questions across academic fields

    • GPT 5: 28.5%
    • GPT Latest: 52.7%Top

  • GPQA Diamond

    Graduate-level science questions

    • GPT 5: 85.4%
    • GPT Latest: 93.9%Top

  • MMLU-Pro

    Multi-subject knowledge and reasoning

    • GPT 5: 87.1%
    • GPT Latest: —

  • IFBench

    Following precise output instructions

    • GPT 5: 73.1%
    • GPT Latest: —

Coding

  • Terminal-Benchv4.0

    Practical software engineering in a terminal

    • GPT 5: —
    • GPT Latest: 49.5%

  • SciCode

    Scientific computing in Python

    • GPT 5: —
    • GPT Latest: 54.2%

  • LiveCodeBench

    Recent competitive programming problems

    • GPT 5: 84.6%
    • GPT Latest: —

Math

  • AIME 2025

    Competition mathematics, 2025 problems

    • GPT 5: 94.3%
    • GPT Latest: —

  • AIME

    Competition mathematics

    • GPT 5: 95.7%
    • GPT Latest: —

  • MATH-500

    Competition math problems

    • GPT 5: 99.4%
    • GPT Latest: —

Legacy evaluations

  • Terminal-Bench Hard

    Earlier terminal task suite

    • GPT 5: 32.6%
    • GPT Latest: —

  • τ²-Bench Telecom

    Tool use in a customer-support simulation

    • GPT 5: 84.8%
    • GPT Latest: —

  • AA-LCR (unversioned)

    Long-context reasoning, earlier grading

    • GPT 5: 78.2%
    • GPT Latest: 79.7%Top

Scores from Artificial Analysis, updated Oct 26, 2026. GPT 5 was tested at high reasoning effort; compare scores measured at the same effort.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": "openai/gpt-5",
    "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="openai/gpt-5",
    messages=[{"role": "user", "content": "Explain this stack trace."}],
)
print(reply.choices[0].message.content)

OpenAI

Other OpenAI models, starting with the same family.