openai

OpenAI: GPT-5.3-Codex

OpenAI's GPT-5.3-Codex is an advanced AI model capable of processing text, images, and files, with a context length of 400,000 tokens and the ability to reason and handle tools, making it versatile for complex tasks. However, structured output support is not available, which might limit its use in specific applications. Given its blended benchmark score of 73.4 across one independent test, this model competes well in general AI tasks. Its pricing at $1.75 per million input tokens and $14 per million output tokens positions it as a higher-cost option compared to some alternatives but offers robust performance for those needing extensive processing capabilities.

Quality Score
100/100
price + capability + benchmarks
Input Price
$1.75
per 1M tokens
Output Price
$14.00
per 1M tokens
Context Window
400,000
tokens

Benchmark results

Independent, published benchmarks. Blended score 73.4 across 1 benchmark, last refreshed 2026-07-23. How scoring works →

BenchmarkMeasuresScore
AI Index broad capability composite 73.0
Model ID
openai/gpt-5.3-codex
Vendor
openai
Released
February 2026
Tokenizer
GPT
Input Modalities
text, image, file
Output Modalities
text
Max Output
128,000 tokens
Tool Calling
✓ supported
Structured Output
✓ supported
Reasoning Mode
✓ supported
Vision
✓ accepts images
Audio
no
Moderated
yes

What it costs in practice

Computed from the current $1.75/M input and $14.00/M output rates. Run your own numbers →

JobTokensCost
Summarize a 50-page report 30k in / 1.5k out $0.07
Classify 1,000 customer emails 500k in / 50k out $1.58
A month of a busy support chatbot 5M in / 2M out $36.75

Similar models

Quick answers

How much does OpenAI: GPT-5.3-Codex cost?
$1.75 per million input tokens and $14.00 per million output tokens.
What is OpenAI: GPT-5.3-Codex's context window?
400,000 tokens, roughly 600 pages of text in a single request.
Does OpenAI: GPT-5.3-Codex support tool calling?
It supports tool calling, structured output, a reasoning mode.
Can OpenAI: GPT-5.3-Codex process images?
Yes, it accepts image input.