Skip to content

GPT-5.5

Released 2026-04 · thinking · 1.0M tokens · 8 benchmarks

Editorial notes

Lanzado 23 abril 2026 (codename Spud). Reasoning model #1 en AA Intelligence Index (60). Lidera Terminal-Bench 2.0 (82.7) — supera a Opus 4.7 por +13 pts. Knowledge cutoff dic 2025. Pricing $5/$30 per MTok (2x sobre 5.4).

Spec sheet

Company
OpenAI
Country
US
Type
thinking
Release
2026-04
Context
1.0M tokens
Slug
gpt-5-5

Benchmarks (8)

Cite this model
BibTeX · APA

BibTeX

@misc{frontier-gpt-5-5,
  title  = {GPT-5.5},
  author = {{OpenAI}},
  year   = {2026},
  note   = {Frontier Benchmarks AI atlas. Accessed 2026-05-08},
  url    = {https://frontierbenchmarks.com/models/gpt-5-5}
}

APA

OpenAI (2026). GPT-5.5 [Large language model]. Frontier Benchmarks AI. Retrieved 2026-05-08, from https://frontierbenchmarks.com/models/gpt-5-5

Citation reflects the atlas page, not the original model paper. For the paper, see the "Resources" section above.