Unlock exclusive introductory pricing for the newly launched Gemini 3.5 Flash.
O

GPT-5.4 nano

Input:$0.16/M
Output:$1/M
Cache Read:$0.016/M
Bแป‘i cแบฃnh:400,000
ฤแบงu ra tแป‘i ฤ‘a:128,000
Ngร y phรกt hร nh:Mar 17, 2026

GPT-5.4 nano is designed for tasks where speed and cost matter most like classification, data extraction, ranking, and sub-agents.

Mแป›i
Phแป• biแบฟn
Sแปญ dแปฅng thฦฐฦกng mแบกi

Playground cho GPT-5.4 nano

Khรกm phรก Playground cแปงa GPT-5.4 nano โ€” mรดi trฦฐแปng tฦฐฦกng tรกc ฤ‘แปƒ kiแปƒm tra mรด hรฌnh vร  chแบกy truy vแบฅn theo thแปi gian thแปฑc. Thแปญ prompts, ฤ‘iแปu chแป‰nh tham sแป‘ vร  lแบทp lแบกi ngay lแบญp tแปฉc ฤ‘แปƒ tฤƒng tแป‘c phรกt triแปƒn vร  xรกc thแปฑc cรกc trฦฐแปng hแปฃp sแปญ dแปฅng.

Technical Specifications of GPT-5.4 Nano

ItemGPT-5.4 Nano (estimated from official + cross-validation)
Model familyGPT-5.4 series (ultra-lightweight โ€œnanoโ€ variant)
ProviderOpenAI
Input typesText
Output typesText
Context window128,000 โ€“ 200,000 tokens (range based on nano tier patterns)
Max output tokens32,000 โ€“ 64,000 tokens (estimated)
Knowledge cutoff~May 31, 2024 (inherited mini/nano lineage)
Reasoning supportLimited (optimized for efficiency over depth)
Tool supportBasic function calling (limited agent capabilities)
PositioningUltra-low-cost, high-throughput inference model

What is GPT-5.4 Nano?

GPT-5.4 Nano is the smallest and most cost-efficient model in the GPT-5.4 family, designed for massive-scale, low-compute workloads. It prioritizes speed, throughput, and cost efficiency over deep reasoning, making it ideal for simple, repeatable tasks.

Unlike GPT-5.4 or GPT-5.4 Mini, Nano is optimized for high-frequency API usage, where millions of requests must be processed quickly and cheaply.

Key Features of GPT-5.4 Nano

  • Ultra-low latency inference: Designed for real-time pipelines and high-QPS systems
  • Extreme cost efficiency: Ideal for large-scale deployments (classification, tagging, routing)
  • Lightweight reasoning: Handles simple instructions reliably but not deep chains
  • High throughput optimization: Built for batch processing and parallel workloads
  • Stable structured output: Works well for JSON formatting, extraction, and labeling tasks
  • Pipeline-friendly design: Commonly used as a โ€œworker modelโ€ in multi-model architectures

Benchmark Performance of GPT-5.4 Nano

  • Not positioned for frontier benchmarks (e.g., SWE-Bench, GPQA)
  • Optimized for:
    • Classification accuracy consistency
    • Structured output reliability
    • Latency benchmarks (substantially faster than Mini/Pro tiers)
  • Typically achieves high precision on narrow tasks but significantly lower performance on reasoning-heavy benchmarks

๐Ÿ‘‰ If you're wondering whether to use the GPT-5.4 Nano or Mini, the key difference is: GPT-5.4 Nano excels in efficiency benchmarks, not reasoning leaderboards.

GPT-5.4-Nano vs Other Models

ModelStrengthContext WindowBest Use Case
GPT-5.4Maximum intelligence~1M tokensComplex reasoning, research
GPT-5.4 MiniBalanced performance + speed~400K tokensCoding, agents
GPT-5.4 NanoFastest + cheapest~400K tokensClassification, extraction
GPT-5 NanoOlder nano baseline~400K tokensBasic NLP tasks

๐Ÿ‘‰ Key takeaway:

  • Use Nano for scale
  • Use Mini for balanced intelligence
  • Use Full/Pro for complex reasoning

Limitations of GPT-5.4 Nano

  • Poor performance on multi-step reasoning or complex logic tasks
  • Limited effectiveness in code generation or advanced analysis
  • Reduced multimodal capability (primarily text-focused)
  • Not suitable for decision-critical or high-accuracy reasoning tasks

Representative Use Cases

  1. Text classification & tagging โ€” sentiment, categories, moderation
  2. Data extraction pipelines โ€” structured JSON output at scale
  3. Routing & orchestration โ€” decide which model/tool to call next
  4. Search indexing & preprocessing โ€” chunk labeling, metadata generation
  5. High-volume automation tasks โ€” millions of lightweight API calls

How to access GPT-5.4 Nano API

Step 1: Sign Up for API Key

Log in toย cometapi.com. If you are not our user yet, please register first. Sign into yourย CometAPI console. Get the access credential API key of the interface. Click โ€œAdd Tokenโ€ at the API token in the personal center, get the token key: sk-xxxxx and submit.

cometapi-key

Step 2: Send Requests to GPT-5.4 Nano API

Select the โ€œgpt-5.4-nanoโ€ endpoint to send the API request and set the request body. The request method and request body are obtained from our website API doc. Our website also provides Apifox test for your convenience. Replace <YOUR_API_KEY> with your actual CometAPI key from your account. base url isย Chat Completionsย andย Responses.

Insert your question or request into the content fieldโ€”this is what the model will respond to . Process the API response to get the generated answer.

Step 3: Retrieve and Verify Results

Process the API response to get the generated answer. After processing, the API responds with the task status and output data.

Cรขu Hแปi Thฦฐแปng Gแบทp

Giรก cแบฃ cho GPT-5.4 nano

Khรกm phรก mแปฉc giรก cแบกnh tranh cho GPT-5.4 nano, ฤ‘ฦฐแปฃc thiแบฟt kแบฟ ฤ‘แปƒ phรน hแปฃp vแป›i nhiแปu ngรขn sรกch vร  nhu cแบงu sแปญ dแปฅng khรกc nhau. Cรกc gรณi linh hoแบกt cแปงa chรบng tรดi ฤ‘แบฃm bแบฃo bแบกn chแป‰ trแบฃ tiแปn cho nhแปฏng gรฌ bแบกn sแปญ dแปฅng, giรบp dแป… dร ng mแปŸ rแป™ng quy mรด khi yรชu cแบงu cแปงa bแบกn tฤƒng lรชn. Khรกm phรก cรกch GPT-5.4 nano cรณ thแปƒ nรขng cao cรกc dแปฑ รกn cแปงa bแบกn trong khi vแบซn kiแปƒm soรกt ฤ‘ฦฐแปฃc chi phรญ.

ModelTierConditionComet Price (USD / M Tokens)Official Price (USD / M Tokens)Discount
gpt-5.4-nanostandard-
Input:$0.1600/M
Output:$1.0000/M
Cache Read:$0.0160/M
Input:$0.2000/M
Output:$1.2500/M
Cache Read:$0.0200/M
-20%
gpt-5.4-nano-2026-03-17standard-
Input:$0.1600/M
Output:$1.0000/M
Cache Read:$0.0160/M
Input:$0.2000/M
Output:$1.2500/M
Cache Read:$0.0200/M
-20%

Mรฃ mแบซu vร  API cho GPT-5.4 nano

Truy cแบญp mรฃ mแบซu toร n diแป‡n vร  tร i nguyรชn API cho GPT-5.4 nano ฤ‘แปƒ tแป‘i ฦฐu hรณa quy trรฌnh tรญch hแปฃp cแปงa bแบกn. Tร i liแป‡u chi tiแบฟt cแปงa chรบng tรดi cung cแบฅp hฦฐแป›ng dแบซn tแปซng bฦฐแป›c, giรบp bแบกn khai thรกc toร n bแป™ tiแปm nฤƒng cแปงa GPT-5.4 nano trong cรกc dแปฑ รกn cแปงa mรฌnh.

#!/usr/bin/env bash
# Get your CometAPI key from https://api.cometapi.com/console/token
# Export it as: export COMETAPI_KEY="your-key-here"

response=$(curl --silent --location --request POST "https://api.cometapi.com/v1/responses" \
  --header "Authorization: Bearer $COMETAPI_KEY" \
  --header "Content-Type: application/json" \
  --header "Accept: application/json" \
  --data-raw '{
    "model": "gpt-5.4-nano",
    "input": "How much gold would it take to coat the Statue of Liberty in a 1mm layer?",
    "reasoning": {
      "effort": "none"
    }
  }')

if command -v jq >/dev/null 2>&1; then
  printf '%s
' "$response" | jq -r '(
    [
      .output[]?
      | select(.type == "message")
      | .content[]?
      | select(.type == "output_text")
      | .text
    ][0]
  ) // .output_text // .'
else
  printf '%s
' "$response"
fi

cURL Code Example

#!/usr/bin/env bash
# Get your CometAPI key from https://api.cometapi.com/console/token
# Export it as: export COMETAPI_KEY="your-key-here"

response=$(curl --silent --location --request POST "https://api.cometapi.com/v1/responses" \
  --header "Authorization: Bearer $COMETAPI_KEY" \
  --header "Content-Type: application/json" \
  --header "Accept: application/json" \
  --data-raw '{
    "model": "gpt-5.4-nano",
    "input": "How much gold would it take to coat the Statue of Liberty in a 1mm layer?",
    "reasoning": {
      "effort": "none"
    }
  }')

if command -v jq >/dev/null 2>&1; then
  printf '%s\n' "$response" | jq -r '(
    [
      .output[]?
      | select(.type == "message")
      | .content[]?
      | select(.type == "output_text")
      | .text
    ][0]
  ) // .output_text // .'
else
  printf '%s\n' "$response"
fi

Python Code Example

from openai import OpenAI
import os

# Get your CometAPI key from https://api.cometapi.com/console/token, and paste it here
COMETAPI_KEY = os.environ.get("COMETAPI_KEY") or "<YOUR_COMETAPI_KEY>"
BASE_URL = "https://api.cometapi.com/v1"

client = OpenAI(base_url=BASE_URL, api_key=COMETAPI_KEY)

response = client.responses.create(
    model="gpt-5.4-nano",
    input="How much gold would it take to coat the Statue of Liberty in a 1mm layer?",
    reasoning={"effort": "none"},
)

print(response.output_text)

JavaScript Code Example

import OpenAI from "openai";

// Get your CometAPI key from https://api.cometapi.com/console/token, and paste it here
const COMETAPI_KEY = process.env.COMETAPI_KEY || "<YOUR_COMETAPI_KEY>";
const BASE_URL = "https://api.cometapi.com/v1";

const client = new OpenAI({
  apiKey: COMETAPI_KEY,
  baseURL: BASE_URL,
});

const response = await client.responses.create({
  model: "gpt-5.4-nano",
  input: "How much gold would it take to coat the Statue of Liberty in a 1mm layer?",
  reasoning: {
    effort: "none",
  },
});

console.log(response.output_text);

Uptime

Tแปท lแป‡ thร nh cรดng cแปงa yรชu cแบงu trong 30 ngร y qua, phแบฃn รกnh ฤ‘แป™ tin cแบญy cแปงa tแปซng nhร  cung cแบฅp mรด hรฌnh. CometAPI giรกm sรกt tแบฅt cแบฃ cรกc nhร  cung cแบฅp ฤ‘ฦฐแปฃc kแบฟt nแป‘i theo thแปi gian thแปฑc, 24/7.

RespondLIVE
1443msAvg. Response
UptimeLIVE
100.0%Avg. Uptime

Cรกc phiรชn bแบฃn cแปงa GPT-5.4 nano

Lรฝ do GPT-5.4 nano cรณ nhiแปu snapshot cรณ thแปƒ bao gแป“m cรกc yแบฟu tแป‘ tiแปm nฤƒng nhฦฐ: sแปฑ thay ฤ‘แป•i ฤ‘แบงu ra sau cรกc bแบฃn cแบญp nhแบญt cแบงn cรกc snapshot cลฉ ฤ‘แปƒ ฤ‘แบฃm bแบฃo tรญnh nhแบฅt quรกn, cung cแบฅp cho nhร  phรกt triแปƒn thแปi gian chuyแปƒn tiแบฟp ฤ‘แปƒ thรญch แปฉng vร  di chuyแปƒn, cลฉng nhฦฐ cรกc snapshot khรกc nhau tฦฐฦกng แปฉng vแป›i cรกc endpoint toร n cแบงu hoแบทc khu vแปฑc nhแบฑm tแป‘i ฦฐu hรณa trแบฃi nghiแป‡m ngฦฐแปi dรนng. ฤแปƒ biแบฟt chi tiแบฟt vแป sแปฑ khรกc biแป‡t giแปฏa cรกc phiรชn bแบฃn, vui lรฒng tham khแบฃo tร i liแป‡u chรญnh thแปฉc.

Version
gpt-5.4-nano
gpt-5.4-nano-2026-03-17