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 modelMiniMax M2.7MiniMaxRemove
ProviderOpenAIMiniMax
Discovery tierFrontierOpen weights
AddedAug 7, 2025Mar 18, 2026
Limits
Context window400KBest204.8K
Max output128K131.07KBest
Average output411 tokens556 tokens
Pricing, USD per 1M tokens
Input$1.25$0.21Best
Output$10.00$0.84Best
Cache read$0.13$0.042Best
Performance
Throughput—35.6 tok/s
Time to first token—0.9 s
Request success (24h)—100.0%
Benchmark indexes
Intelligence Index23.0Best22.8
Coding Index37.852.6Best
Math Index94.3—
Capabilities
ReasoningYesYes
ToolsYesYes
VisionYesNo
Structured outputYesYes
PDF inputYesNo
Reasoning effortminimal, low, medium, high—

Public evaluations

Benchmarks

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

  • GPT 5
  • MiniMax M2.7

Intelligence Index

  • GPT 523.0
  • MiniMax M2.722.8

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
    • MiniMax M2.7: 52.6Top

  • Math Index

    Composite of the math evaluations

    • GPT 5: 94.3
    • MiniMax M2.7: —

Agentic work

  • AutomationBench-AA

    Workflow automation, penalized for guardrail violations

    • GPT 5: —
    • MiniMax M2.7: 4.4%

  • AutomationBench-AA tasks completed

    Workflows finished end to end without a guardrail violation

    • GPT 5: —
    • MiniMax M2.7: 0.3%

  • AA-Briefcasev1.1

    Agentic knowledge work, rated head to head

    • GPT 5: —
    • MiniMax M2.7: 718 Elo

  • GDPval-AAv2.1

    Economically valuable occupational tasks, rated head to head

    • GPT 5: 918 Elo
    • MiniMax M2.7: 1,008 EloTop

Document reasoning

  • GDP.pdf

    Professional reasoning over PDF documents, all-pass rate

    • GPT 5: —
    • MiniMax M2.7: 6.0%

  • AA-LCRv1.1

    Reasoning across long documents

    • GPT 5: 78.2%
    • MiniMax M2.7: 78.3%Top

  • MLCR-AA

    Long-context reasoning over medical records

    • GPT 5: —
    • MiniMax M2.7: 5.6%

Reasoning and knowledge

  • Humanity’s Last Exam

    Expert-level questions across academic fields

    • GPT 5: 28.5%
    • MiniMax M2.7: 29.6%Top

  • GPQA Diamond

    Graduate-level science questions

    • GPT 5: 85.4%
    • MiniMax M2.7: 87.4%Top

  • MMLU-Pro

    Multi-subject knowledge and reasoning

    • GPT 5: 87.1%
    • MiniMax M2.7: —

  • IFBench

    Following precise output instructions

    • GPT 5: 73.1%
    • MiniMax M2.7: 75.7%Top

Coding

  • Terminal-Benchv4.0

    Practical software engineering in a terminal

    • GPT 5: —
    • MiniMax M2.7: 0.0%

  • SciCode

    Scientific computing in Python

    • GPT 5: —
    • MiniMax M2.7: 50.1%

  • LiveCodeBench

    Recent competitive programming problems

    • GPT 5: 84.6%
    • MiniMax M2.7: —

Math

  • AIME 2025

    Competition mathematics, 2025 problems

    • GPT 5: 94.3%
    • MiniMax M2.7: —

  • AIME

    Competition mathematics

    • GPT 5: 95.7%
    • MiniMax M2.7: —

  • MATH-500

    Competition math problems

    • GPT 5: 99.4%
    • MiniMax M2.7: —

Legacy evaluations

  • Terminal-Bench Hard

    Earlier terminal task suite

    • GPT 5: 32.6%
    • MiniMax M2.7: 39.4%Top

  • τ²-Bench Telecom

    Tool use in a customer-support simulation

    • GPT 5: 84.8%Top
    • MiniMax M2.7: 84.8%Top

  • AA-LCR (unversioned)

    Long-context reasoning, earlier grading

    • GPT 5: 78.2%
    • MiniMax M2.7: 78.3%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.