GPT-5.3 Chat

Provider: OpenAI US

No longer listed on OpenRouter since 2026-08-10. The figures below are the last recorded values.
Input price
$1.75
/ 1M tokens
Output price
$14.00
/ 1M tokens
Cache price
$0.17
/ 1M tokens (read)
Context
128K tokens
Max output 16K tokens

Overview

OpenAI's GPT-5.3 Chat is positioned as a flagship model, targeting complex conversational and multi-modal workloads. It offers advanced capabilities like vision and structured output modes, suitable for demanding applications.

Strengths

  • Supports advanced multi-modal interaction with integrated vision capabilities.
  • Enables robust tool integration through function-calling support.
  • Delivers precise, structured responses via dedicated JSON mode.
  • Provides a substantial 128,000 token context window for extensive input.
  • Features a generous maximum output token limit of 16,384 tokens.

Best use cases

  • Building intelligent assistants that process visual and textual inputs.
  • Developing sophisticated agents capable of interacting with external tools.
  • Generating structured data or code snippets in reliable JSON format.
  • Analyzing large documents or conversations requiring extensive context.
  • Implementing interactive applications where both input and output can be lengthy.

Pricing position

GPT-5.3 Chat is positioned as OpenAI's premium offering, with an input price of $1.75/1M tokens and an output price of $14/1M tokens. This makes it significantly more expensive than all listed OpenAI sibling models like GPT-5 Mini ($0.25/M in, $2/M out). Among similarly-priced peers, its input price is competitive, but its $14/1M output price is the highest, notably above Mistral Large 2411 ($6/M out) and Sonar Reasoning Pro ($8/M out).

Caveats

Developers should note the absence of a free tier for GPT-5.3 Chat, necessitating upfront cost considerations. While powerful, its high output price of $14/1M tokens might be prohibitive for high-volume, output-heavy applications where cheaper alternatives like Mistral Medium 3.5 ($7.5/M out) or even OpenAI's own GPT-5 Mini ($2/M out) might suffice. For extremely long context tasks without needing vision or function-calling, GPT-4.1 Nano offers a much larger 1,047,576 token context at a fraction of the cost.

Notes drafted by an LLM from the figures on this page (2026-05-21) and rewritten when the price changes.

Key facts

Max output
16K tokens
Tracked since
2026-05-08
Last updated
2026-08-10

Features

Image inputFunction callingJSON output

Price history

  • 2026-08-10 No longer listed

Monthly cost estimate

Scale Input (M) Output (M) Monthly
0.1M (hobby) 0.1 0.05 $0.88
1M (small app) 1 0.5 $8.75
10M (mid scale) 10 5 $87.50
100M (large) 100 50 $875.00

※ Cache and batch discounts not factored in. Detailed calculator

Official resources

Other models from OpenAI

Similar-priced models

Source: OpenRouter API