GPT-5.2

reasoningproprietary

GPT-5.2 is our previous flagship model for complex professional work. We recommend using the latest GPT-6 Astra. Learn more on the Model guidance page. Reasoning.effort supports: none (default), low, medium, high and xhigh.

Intelligence
Reasoning
Speed
Context
400K
Max output
128K
Input price
$1.75/1M tokens
Output price
$14/1M tokens
Snapshotsgpt-5.2-2025-12-11

Capabilities

Reasoning & Interaction
Reasoning
Tool Calling
Output Control
Structured Output
JSON Mode
Infrastructure
Streaming
Batch Processing

Details

Provider OpenAI
Creator openai
Familygpt-5.2
Licenseproprietary
Statusactive
Typereasoning
Input modalitiestext, image
Output modalitiestext
Open weightNo
Parameters—
Architecture—
Context window400K tokens
Max input—
Max output128K tokens
Reasoning tokensYes
Knowledge cutoff
Release date
Sourceofficial

Tools

Function Callingfunction_calling
Call external functions and APIs
Web Searchweb_search
Search the web for information
File Searchfile_search
Search across uploaded files
Image Generationimage_generation
Generate images inline
Code Interpretercode_interpreter
Execute code in a sandbox
Hosted Shellhosted_shell
Run shell commands in a container
Apply Patchapply_patch
Apply code patches to files
Skillsskills
Use built-in skill modules
MCPmcp
Connect to Model Context Protocol servers

Endpoints

Chat CompletionsPOST
Generate chat responses with messageshttps://api.openai.com/v1/chat/completions
ResponsesPOST
Stateful multi-turn conversationshttps://api.openai.com/v1/responses
BatchPOST
Async batch processing at lower costhttps://api.openai.com/v1/batches

Pricing

Text tokensPer 1M tokens
InputCached inputOutput
Standard$1.75$0.175$14

Family Comparison: gpt-5.2

ModelContextMax outPricing
GPT-5.2 Chat128K16K$1.75$14
GPT-5.2-Codex400K128K$1.75$14
GPT-5.2400K128K$1.75$14
GPT-5.2 Pro400K128K$21$168

API

GET/v1/models/openai/gpt-5.2