BrowseComp Long Context 256k Benchmark Leaderboard | Phaseo

BrowseComp Long Context 256k

Type: percentage

Recorded Results

6

Average Score

69.90%

Score Range

19.10% - 88.80%

Leading Model (lowest score)

19.10% - GPT 4.1 Nano

Scores Over Time

Individual benchmark scores plotted by date.

Models Using This Benchmark

Organisation	Model	Reported	Top Score	Info	Self Reported	Source
OpenAI	GPT 4.1 Nano	14 Apr 2025	19.10%	-	Yes	Source
OpenAI	GPT 5 Nano	07 Aug 2025	68.40%	High Reasoning Effort	Yes	Source
OpenAI	GPT 4.1	14 Apr 2025	75.50%	-	Yes	Source
OpenAI	GPT 4.1 Mini	14 Apr 2025	81.60%	-	Yes	Source
OpenAI	GPT 5 Mini	07 Aug 2025	86%	High Reasoning Effort	Yes	Source
OpenAI	GPT 5	07 Aug 2025	88.80%	High Reasoning Effort	Yes	Source