NOUMENOpen app
NOUMEN

OpenAI · Efficient

GPT-4o mini

Added Jan 1, 2024

Compare models

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

Loading models…

Use in the app

OpenAI's most cost-efficient small model. Cheaper and smarter than GPT-3.5 (the original ChatGPT), but less performant than gpt-4o.

  • Vision
Input
Text, Images
Output
Text
Discovery tier
Efficient

Pricing

USD per 1M tokens
Input
$0.15
Output
$0.60
Cache read
$0.075

Reference prices as published by the model provider. API example

Specifications

Context window
128K
Max output
16.4K
Average output (7 days)
29 tokens
Longer than 1% of 465 models

Side by side

GPT-4o mini 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-4o miniOpenAIThis modelGPT 6 LunaOpenAIRemove
ProviderOpenAIOpenAI
Discovery tierEfficientEfficient
AddedJan 1, 2024Sep 22, 2026
Limits
Context window128K1.05MBest
Max output16.38K128KBest
Average output29 tokens120 tokens
Pricing, USD per 1M tokens
Input$0.15$0.10Best
Output$0.60$0.50Best
Cache read$0.075$0.01Best
Cache write—$0.13
Performance
Throughput—31.2 tok/s
Time to first token—7.5 s
Request success (24h)100.0%Best99.8%
Benchmark indexes
Intelligence Index6.729.9Best
Coding Index11.4—
Math Index14.7—
Capabilities
ReasoningNoYes
ToolsNoYes
VisionYesYes
Structured outputNoYes
PDF inputNoYes
Reasoning effort—none, low, medium, high, xhigh, max

Public evaluations

Benchmarks

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

  • GPT-4o mini
  • GPT 6 Luna

Intelligence Index

  • GPT-4o mini6.7
  • GPT 6 Luna29.9

GPT-4o mini: Better than 6% of 326 scored models

3.8Catalog models by score55.1

Indexes

  • Coding Index

    Composite of the coding evaluations

    • GPT-4o mini: 11.4
    • GPT 6 Luna: —

  • Math Index

    Composite of the math evaluations

    • GPT-4o mini: 14.7
    • GPT 6 Luna: —

Agentic work

  • AutomationBench-AA

    Workflow automation, penalized for guardrail violations

    • GPT-4o mini: —
    • GPT 6 Luna: 40.5%

  • AutomationBench-AA tasks completed

    Workflows finished end to end without a guardrail violation

    • GPT-4o mini: —
    • GPT 6 Luna: 10.2%

  • AA-Briefcasev1.1

    Agentic knowledge work, rated head to head

    • GPT-4o mini: —
    • GPT 6 Luna: 1,097 Elo

  • GDPval-AAv2.1

    Economically valuable occupational tasks, rated head to head

    • GPT-4o mini: —
    • GPT 6 Luna: 1,263 Elo

Document reasoning

  • GDP.pdf

    Professional reasoning over PDF documents, all-pass rate

    • GPT-4o mini: —
    • GPT 6 Luna: 13.8%

  • AA-LCRv1.1

    Reasoning across long documents

    • GPT-4o mini: —
    • GPT 6 Luna: 78.3%

  • MLCR-AA

    Long-context reasoning over medical records

    • GPT-4o mini: —
    • GPT 6 Luna: 18.3%

Reasoning and knowledge

  • Humanity’s Last Exam

    Expert-level questions across academic fields

    • GPT-4o mini: 4.2%
    • GPT 6 Luna: 28.3%Top

  • GPQA Diamond

    Graduate-level science questions

    • GPT-4o mini: 42.6%
    • GPT 6 Luna: —

  • MMLU-Pro

    Multi-subject knowledge and reasoning

    • GPT-4o mini: 64.8%
    • GPT 6 Luna: —

  • IFBench

    Following precise output instructions

    • GPT-4o mini: 31.0%
    • GPT 6 Luna: —

Coding

  • Terminal-Benchv4.0

    Practical software engineering in a terminal

    • GPT-4o mini: —
    • GPT 6 Luna: 2.5%

  • SciCode

    Scientific computing in Python

    • GPT-4o mini: —
    • GPT 6 Luna: 50.9%

  • LiveCodeBench

    Recent competitive programming problems

    • GPT-4o mini: 23.4%
    • GPT 6 Luna: —

Math

  • AIME 2025

    Competition mathematics, 2025 problems

    • GPT-4o mini: 14.7%
    • GPT 6 Luna: —

  • AIME

    Competition mathematics

    • GPT-4o mini: 11.7%
    • GPT 6 Luna: —

  • MATH-500

    Competition math problems

    • GPT-4o mini: 78.9%
    • GPT 6 Luna: —

Legacy evaluations

  • AA-LCR (unversioned)

    Long-context reasoning, earlier grading

    • GPT-4o mini: —
    • GPT 6 Luna: 78.3%

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.

Prompt cache hits
0.0%
About $0.15 per 1M input tokens with caching
Request success (24h)
100.0%
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 100.0%
  • 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": "openai/gpt-4o-mini",
    "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-4o-mini",
    messages=[{"role": "user", "content": "Explain this stack trace."}],
)
print(reply.choices[0].message.content)

OpenAI

Other OpenAI models, starting with the same family.