NOUMENOpen app
NOUMEN

OpenAI · Efficient

GPT 5.4 Mini

Added Mar 17, 2026

Compare models

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

Loading models…

Use in the app

GPT-5.4 mini brings the core capabilities of GPT-5.4 to a faster, more efficient model optimized for high-throughput workloads.

  • Reasoning
  • Tools
  • Vision
  • Structured output
  • PDF input
Input
Text, Images
Output
Text
Reasoning effort
none · low · medium · high · xhigh
Discovery tier
Efficient

Pricing

USD per 1M tokens
Input
$0.75
Output
$4.50
Cache read
$0.075

Reference prices as published by the model provider. API example

Specifications

Context window
400K
Max output
128K
Average output (7 days)
172 tokens
Longer than 17% of 465 models

Side by side

GPT 5.4 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 5.4 MiniOpenAIThis modelGPT 6.1 SolOpenAIRemove
ProviderOpenAIOpenAI
Discovery tierEfficientFrontier
AddedMar 17, 2026Sep 29, 2026
Limits
Context window400K1.05MBest
Max output128K128K
Average output172 tokens388 tokens
Pricing, USD per 1M tokens
Input$0.75Best$2.00
Output$4.50Best$10.00
Cache read$0.075Best$0.10
Cache write—$2.50
Performance
Throughput128.9 tok/sBest115.7 tok/s
Time to first token0.8 sBest3.6 s
Request success (24h)100.0%Best99.9%
Benchmark indexes
Intelligence Index11.147.8Best
Capabilities
ReasoningYesYes
ToolsYesYes
VisionYesYes
Structured outputYesYes
PDF inputYesYes
Reasoning effortnone, low, medium, high, xhighlow, medium, high, xhigh, max

Public evaluations

Benchmarks

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

  • GPT 5.4 Mini
  • GPT 6.1 Sol

Intelligence Index

  • GPT 5.4 Mini11.1
  • GPT 6.1 Sol47.8

GPT 5.4 Mini: Better than 23% of 326 scored models

3.8Catalog models by score55.1

Agentic work

  • AutomationBench-AA

    Workflow automation, penalized for guardrail violations

    • GPT 5.4 Mini: —
    • GPT 6.1 Sol: 62.6%

  • AutomationBench-AA tasks completed

    Workflows finished end to end without a guardrail violation

    • GPT 5.4 Mini: —
    • GPT 6.1 Sol: 30.1%

  • AA-Briefcasev1.1

    Agentic knowledge work, rated head to head

    • GPT 5.4 Mini: —
    • GPT 6.1 Sol: 1,359 Elo

  • GDPval-AAv2.1

    Economically valuable occupational tasks, rated head to head

    • GPT 5.4 Mini: 589 Elo
    • GPT 6.1 Sol: 1,450 EloTop

Document reasoning

  • GDP.pdf

    Professional reasoning over PDF documents, all-pass rate

    • GPT 5.4 Mini: —
    • GPT 6.1 Sol: 30.0%

  • AA-LCRv1.1

    Reasoning across long documents

    • GPT 5.4 Mini: 37.0%
    • GPT 6.1 Sol: 83.3%Top

  • MLCR-AA

    Long-context reasoning over medical records

    • GPT 5.4 Mini: —
    • GPT 6.1 Sol: 32.2%

Reasoning and knowledge

  • Humanity’s Last Exam

    Expert-level questions across academic fields

    • GPT 5.4 Mini: 5.9%
    • GPT 6.1 Sol: 49.9%Top

  • GPQA Diamond

    Graduate-level science questions

    • GPT 5.4 Mini: 60.6%
    • GPT 6.1 Sol: —

  • IFBench

    Following precise output instructions

    • GPT 5.4 Mini: 38.8%
    • GPT 6.1 Sol: —

Coding

  • Terminal-Benchv4.0

    Practical software engineering in a terminal

    • GPT 5.4 Mini: —
    • GPT 6.1 Sol: 48.0%

  • SciCode

    Scientific computing in Python

    • GPT 5.4 Mini: —
    • GPT 6.1 Sol: 53.2%

Legacy evaluations

  • Terminal-Bench Hard

    Earlier terminal task suite

    • GPT 5.4 Mini: 18.2%
    • GPT 6.1 Sol: —

  • τ²-Bench Telecom

    Tool use in a customer-support simulation

    • GPT 5.4 Mini: 23.4%
    • GPT 6.1 Sol: —

  • AA-LCR (unversioned)

    Long-context reasoning, earlier grading

    • GPT 5.4 Mini: 37.0%
    • GPT 6.1 Sol: 83.3%Top

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
128.9 tok/s
Faster than 72% of 251 models
Time to first token
0.8 s
Quicker than 97% of 251 models
Prompt cache hits
29.1%
About $0.55 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-5.4-mini",
    "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-5.4-mini",
    messages=[{"role": "user", "content": "Explain this stack trace."}],
)
print(reply.choices[0].message.content)

OpenAI

Other OpenAI models, starting with the same family.