← Models

GPT-5.4 nano

OpenAI · GPT-5

OpenAI's cheapest GPT-5.4 route for fast classification, extraction, and lightweight coding subagents.

Part of GPT family · Other versions: GPT-5.6 Luna , GPT-5.6 Terra , GPT-5.6 Sol , GPT-5.5 Pro , GPT-5.5 , GPT-Rosalind , GPT-5.4 mini , GPT-5.4 Pro , GPT-5.4 , GPT-5.3 , GPT-5.3-Codex , GPT-5.2-Codex , GPT-5.2 Pro , GPT-5.2 , GPT-5-Codex , GPT-5 mini , GPT-5 nano , GPT-5 , GPT-4.1 , GPT-4o mini Transcribe , GPT-4o mini TTS , GPT-4o Transcribe , Computer Use Preview , GPT-4o mini , GPT-4o
Type
language
Context
400K tokens
Max Output
128K tokens
Status
current
Input
$0.2/1M tok
Output
$1.25/1M tok
API Access
Yes
License
proprietary
classification extraction subagents low-cost multimodal automation
Released March 2026 · Updated June 8, 2026

Overview

Freshness note: Model capabilities, limits, and pricing can change quickly. This profile is a point-in-time snapshot last verified on June 8, 2026.

GPT-5.4 nano is OpenAI’s smallest GPT-5.4 model, launched alongside GPT-5.4 mini on March 17, 2026. It is intended for high-throughput supporting tasks where teams want the newest GPT-5.4-era quality improvements at the lowest possible cost.

Capabilities

GPT-5.4 nano is best for classification, extraction, ranking, lightweight coding help, and other bounded subagent tasks that do not justify a larger reasoning model. OpenAI explicitly recommends it for simpler supporting work inside multi-model systems, especially when fast turnaround matters more than deep open-ended analysis.

Technical Details

OpenAI’s current model docs list GPT-5.4 nano with a 400K context window and 128K max output. It accepts text and image input, returns text output, and supports streaming, function calling, structured outputs, web search, file search, image generation, code interpreter, hosted shell, apply patch, skills, and MCP. The caveat is that current docs list computer use and tool search as not supported for nano.

That matters because GPT-5.4 nano is not just a cheap classifier. It can still participate in broader tool-based workflows, which makes it useful as the low-cost worker tier beneath larger coordinator models.

Pricing & Access

Published API pricing is:

  • Input: $0.20 per 1M tokens
  • Cached input: $0.02 per 1M tokens
  • Output: $1.25 per 1M tokens

OpenAI also notes a 10% uplift for regional-processing endpoints. Unlike GPT-5.4 mini, GPT-5.4 nano is primarily a builder-facing model rather than a general ChatGPT-facing route.

Best Use Cases

Use GPT-5.4 nano for high-volume event labeling, document extraction, support-ticket triage, ranking, enrichment, and narrow coding subtasks delegated from a larger model. It is a strong fit when a system needs many cheap, parallel support calls with better reliability than the older GPT-5 nano tier.

Comparisons

  • GPT-5.4 mini (OpenAI): Better for harder coding, computer use, and agentic execution, but materially more expensive.
  • GPT-5 nano (OpenAI): Older ultra-cheap baseline that OpenAI now positions below this newer nano tier.
  • Gemini 2.5 Flash-Lite (Google): Comparable category for high-frequency lower-cost work, with tradeoffs mostly driven by ecosystem and multimodal tooling fit.