BrowseComp Long Context 128k Benchmark Leaderboard | Phaseo

BrowseComp Long Context 128k

Type: percentage

Recorded Results

8

Average Score

86.55%

Score Range

80% - 90%

Leading Model (lowest score)

80% - o4 Mini

Scores Over Time

Individual benchmark scores plotted by date.

Models Using This Benchmark

Organisation	Model	Reported	Top Score	Info	Self Reported	Source
OpenAI	o4 Mini	16 Apr 2025	80%	High Reasoning Effort	Yes	Source
OpenAI	GPT 5 Nano	07 Aug 2025	80.40%	High Reasoning Effort	Yes	Source
OpenAI	GPT 4.1	14 Apr 2025	85.90%	-	Yes	Source
OpenAI	o3	16 Apr 2025	88.30%	High Reasoning Effort	Yes	Source
OpenAI	GPT 4.1 Mini	14 Apr 2025	89%	-	Yes	Source
OpenAI	GPT 4.1 Nano	14 Apr 2025	89.40%	-	Yes	Source
OpenAI	GPT 5 Mini	07 Aug 2025	89.40%	High Reasoning Effort	Yes	Source
OpenAI	GPT 5	07 Aug 2025	90%	High Reasoning Effort	Yes	Source