SYLIN

GPT

by OpenAI · Model · Language & code models

promising

GPT-5.5 reads documents well, but ranks third behind two Claude models

Took #1 on Arena document among open weights

Why we say so

Arena's document-understanding ranking puts GPT-5.5 third at 1484, behind Claude Opus 5 (1509) and Claude Fable 5.1 (1515). That is one independent ranking, and the gap to first place is real but modest. Nothing here covers GPT 5.4 or hands-on use.

Versions

VersionReleasedRank
GPT-5.5LATEST
promising Third on Arena document understanding, behind two Claude models
#3 documents
GPT-5.4

Evidence

  1. Nov 9
    #3 on Arena document (1484)
  2. Sep 13
    #3 on Arena document (1486)

Alternatives

Claude Opus#2 on Arena document (1509)
Claude Fable#1 on Arena document (1515)
Intern-SLeads open-source models on MMLU Pro and SimpleQA Verified, but trails GPT-5.5 and Claude Opus 4.8 on Terminal-Bench and SWE-bench Pro
GPT SolSeveral users find 5.6 Sol over-engineers simple tasks and is slow versus 5.4, e.g. a four-page answer for a todo list
QwenIn an earlier experiment Qwen barely moved toward Opus 4.8 when prefilled with its reasoning, unlike its shift toward GPT-5.5 Pro