ModelRefs / BrowseComp Long Context Leaderboard
BrowseComp Long Context Leaderboard
OpenAI long-context question-answering benchmark with relevant search results embedded in inputs up to hundreds of thousands of tokens.
Overview
How it is measured: Exact-answer accuracy over grounded long-context question-answering tasks, reported separately by input-length range.
How this benchmark is scored
| Category | retrieval |
|---|---|
| Maximum score | 100 % accuracy |
| Direction | Higher is better |
Primary source: https://huggingface.co/datasets/openai/browsecomp-long-context
Continue your research
Use these connected ModelRefs sections to compare alternatives, inspect implementation paths, and review the evidence and governance boundaries relevant to BrowseComp Long Context Leaderboard.
Frequently asked questions
What does the BrowseComp Long Context benchmark measure?
OpenAI long-context question-answering benchmark with relevant search results embedded in inputs up to hundreds of thousands of tokens.
What is the methodology?
Exact-answer accuracy over grounded long-context question-answering tasks, reported separately by input-length range.
Which AI model is the current BrowseComp Long Context leader?
GPT-5 (OpenAI) currently leads with 90 % accuracy.