NOUMENOpen app
NOUMEN

OpenAI · Efficient

GPT 5.4 Nano

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 nano is the most lightweight and cost-efficient variant of the GPT-5.4 family, optimized for speed-critical and high-volume tasks.

  • 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
$0.20
Output
$1.25
Cache read
$0.02

Reference prices as published by the model provider. API example

Specifications

Context window
400K
Max output
128K
Average output (7 days)
515 tokens
Longer than 50% of 468 models

Side by side

GPT 5.4 Nano 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 NanoOpenAIThis modelMistral Small 4 119B ThinkingMistralRemove
ProviderOpenAIMistral
Discovery tierEfficientOpen weights
AddedMar 17, 2026Mar 17, 2026
Limits
Context window400KBest262.14K
Max output128KBest16.38K
Average output515 tokens997 tokens
Pricing, USD per 1M tokens
Input$0.20Best$0.40
Output$1.25Best$1.40
Cache read$0.02—
Performance
Throughput—123.6 tok/s
Time to first token—1.7 s
Request success (24h)100.0%Best95.9%
Benchmark indexes
Intelligence Index11.7Best11.3
Coding Index—26.6
Capabilities
ReasoningYesYes
ToolsYesYes
VisionYesYes
Structured outputYesYes
PDF inputYesNo
Reasoning effortnone, low, medium, high, xhighnone, high

Public evaluations

Benchmarks

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

  • GPT 5.4 Nano
  • Mistral Small 4 119B Thinking

Intelligence Index

  • GPT 5.4 Nano11.7
  • Mistral Small 4 119B Thinking11.3

GPT 5.4 Nano: Better than 26% of 326 scored models

3.8Catalog models by score55.1

Indexes

  • Coding Index

    Composite of the coding evaluations

    • GPT 5.4 Nano: —
    • Mistral Small 4 119B Thinking: 26.6

Agentic work

  • AutomationBench-AA

    Workflow automation, penalized for guardrail violations

    • GPT 5.4 Nano: —
    • Mistral Small 4 119B Thinking: 1.0%

  • AutomationBench-AA tasks completed

    Workflows finished end to end without a guardrail violation

    • GPT 5.4 Nano: —
    • Mistral Small 4 119B Thinking: 0.0%

  • AA-Briefcasev1.1

    Agentic knowledge work, rated head to head

    • GPT 5.4 Nano: —
    • Mistral Small 4 119B Thinking: 241 Elo

  • GDPval-AAv2.1

    Economically valuable occupational tasks, rated head to head

    • GPT 5.4 Nano: —
    • Mistral Small 4 119B Thinking: 362 Elo

Document reasoning

  • GDP.pdf

    Professional reasoning over PDF documents, all-pass rate

    • GPT 5.4 Nano: —
    • Mistral Small 4 119B Thinking: 0.0%

  • AA-LCRv1.1

    Reasoning across long documents

    • GPT 5.4 Nano: 29.7%
    • Mistral Small 4 119B Thinking: 49.7%Top

  • MLCR-AA

    Long-context reasoning over medical records

    • GPT 5.4 Nano: —
    • Mistral Small 4 119B Thinking: 1.7%

Reasoning and knowledge

  • Humanity’s Last Exam

    Expert-level questions across academic fields

    • GPT 5.4 Nano: 4.1%
    • Mistral Small 4 119B Thinking: 9.9%Top

  • GPQA Diamond

    Graduate-level science questions

    • GPT 5.4 Nano: 55.8%
    • Mistral Small 4 119B Thinking: 76.9%Top

  • IFBench

    Following precise output instructions

    • GPT 5.4 Nano: 32.7%
    • Mistral Small 4 119B Thinking: 48.2%Top

Coding

  • Terminal-Benchv4.0

    Practical software engineering in a terminal

    • GPT 5.4 Nano: —
    • Mistral Small 4 119B Thinking: 0.0%

  • SciCode

    Scientific computing in Python

    • GPT 5.4 Nano: —
    • Mistral Small 4 119B Thinking: 38.8%

Legacy evaluations

  • Terminal-Bench Hard

    Earlier terminal task suite

    • GPT 5.4 Nano: 24.2%Top
    • Mistral Small 4 119B Thinking: 17.4%

  • τ²-Bench Telecom

    Tool use in a customer-support simulation

    • GPT 5.4 Nano: 34.8%
    • Mistral Small 4 119B Thinking: 41.2%Top

  • AA-LCR (unversioned)

    Long-context reasoning, earlier grading

    • GPT 5.4 Nano: 29.7%
    • Mistral Small 4 119B Thinking: 49.7%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.

Request success (24h)
100.0%
Reliable
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-nano",
    "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-nano",
    messages=[{"role": "user", "content": "Explain this stack trace."}],
)
print(reply.choices[0].message.content)

OpenAI

Other OpenAI models, starting with the same family.