GPT-5.4 nano

OpenAIMCP Tool CallReasoning

openai/gpt-5.4-nanoClosed API

Cheapest GPT-5.4 lane for simple routing, extraction, and bulk automation

Benchmark Pricing$0.20 / 1M in$1.25 / 1M out$0.02 cache read

Context: 400,000 tokens•Max Output: 128,000 tokens•Knowledge: 2025-08-31•Released: 2026-03-17

TECHNICAL SPECIFICATIONS

Context Window400,000400K tokens

Max Output128,000128K tokens

LicenseProprietaryCloud API only

Input Modalities

TextVision

Function / Tool CallingSupported (MCP Ready)

Structured OutputsGuaranteed Schema

MODEL CONTEXT PROTOCOL (MCP) READINESS

✓ GPT-5.4 nano natively supports OpenAI / Anthropic format tool calling and is fully compatible with MCP servers and agent loops.

MCP Context Window Recommendations

With a context capacity of 400,000 tokens, this model can handle moderately large MCP tool definitions and multi-turn conversation logs comfortably.

STANDARDIZED BENCHMARKS & EVALS

SWE-Bench Proresolve rate

52.4

Terminal-Benchaccuracy

46.3

MCP Atlasscore

56.1

Toolathlonscore

35.5

τ²-Bench Telecomaccuracy

92.5

GPQA Diamondaccuracy

82.8

Humanity's Last Examaccuracy

37.7

Humanity's Last Examaccuracy

24.3

OSWorld-Verifiedsuccess rate

MMMU Proaccuracy

69.5

MMMU Proaccuracy

66.1

OmniDocBenchoverall edit distance

0.2419

OpenAI MRCRaccuracy

44.2

OpenAI MRCRaccuracy

33.1

Graphwalksaccuracy

73.4

Graphwalksaccuracy

50.8

API CALL INVOCATION EXAMPLE

// Example: Calling GPT-5.4 nano via OpenAI SDK
import OpenAI from "openai";

const client = new OpenAI({
  apiKey: process.env.OPENAI_API_KEY,
});

const response = await client.chat.completions.create({
  model: "openai/gpt-5.4-nano",
  messages: [{ role: "user", content: "Hello! Can you inspect MCP tools?" }],
});

console.log(response.choices[0].message.content);