← All models

GPT-5.5

Published Publisher OpenAIgpt-5.5

Positioning

GPT-5.5 is OpenAI’s flagship reasoning model for complex professional work, including coding and multi-step tasks involving tools. The API alias is gpt-5.5; its documented default snapshot is gpt-5.5-2026-04-23. (openai.com)

Capabilities and uses

  • Writes and debugs code, researches online, analyzes data, creates documents and spreadsheets, and helps with scientific research and other knowledge work.
  • Accepts text and image input and produces text output. It supports function calling, structured outputs and, through the Responses API, tools including web search, code interpreter and computer use.
  • The documented context window is 1,050,000 tokens, with up to 128,000 output tokens. Reasoning effort can be set to none, low, medium, high or xhigh; the default is medium. (openai.com)

Limitations

The model’s documented knowledge cutoff is December 1, 2025. Audio and video input are not supported, and fine-tuning is not supported. Image generation is available through a Responses API tool rather than as native image output from this model. Results and tool actions should be checked before reliance, particularly for consequential work. GPT-5.5 prompt caching requires extended prompt caching; in-memory caching is not supported. (developers.openai.com)

Release

OpenAI announced GPT-5.5 on April 23, 2026. Its announcement was updated on April 24, 2026 to state that GPT-5.5 was available in the API. (openai.com)

API pricing

OpenAI API text-token prices, in USD per 1 million tokens, checked November 16, 2026. These are for gpt-5.5, not GPT-5.5 Pro. (developers.openai.com)

Processing Input tokens Input Cached input Output
Standard ≤272,000 $5.00 $0.50 $30.00
Standard >272,000 $10.00 $1.00 $45.00
Batch ≤272,000 $2.50 $0.25 $15.00
Batch >272,000 $5.00 $0.50 $22.50
Flex ≤272,000 $2.50 $0.25 $15.00
Flex >272,000 $5.00 $0.50 $22.50
Fast ≤272,000 $12.50 $1.25 $75.00

For prompts exceeding 272,000 input tokens, the long-context rates apply to the full session for Standard, Batch and Flex. Fast-mode long-context pricing was not verified. Regional processing endpoints for data residency carry a 10% uplift. Tool calls can incur separate charges. (developers.openai.com)

Official sources