reasoning benchmark
BrowseComp Long Context 128k
A challenging benchmark for evaluating web browsing agents' ability to persistently navigate the internet and find hard-to-locate, entangled information. Comprises 1,266 questions requiring strategic reasoning, creative search, and interpretation of retrieved content, with short and easily verifiable answers.
Updated Aug 7, 2026
Published models5
Registry coverage5
MetricScore
EvidenceB
BrowseComp Long Context 128k leaderboard
Sorted by the source-provided rank. Higher score is better according to the registry.
| 1 | OP | 92.0% | 100.0% | 5 | C | |
| 2 | OP | 90.0% | 75.0% | 5 | C | |
| 3 | OP | 90.0% | 50.0% | 5 | C | |
| 4 | OP | 90.0% | 25.0% | 5 | C | |
| 5 | OP | 90.0% | 0.0% | 5 | C |
Score distribution
Top published rows on the benchmark's original scale.
BrowseComp Long Context 128k highlights
The top published results on this benchmark's own scale.
What is BrowseComp Long Context 128k?
Definition and scoring fields from the benchmark registry.
A challenging benchmark for evaluating web browsing agents' ability to persistently navigate the internet and find hard-to-locate, entangled information. Comprises 1,266 questions requiring strategic reasoning, creative search, and interpretation of retrieved content, with short and easily verifiable answers.
Scores are shown in ratio. The current registry marks this benchmark as not independently verified with evidence level B.
- Family
- BrowseComp Long Context 128k
- Modality
- text
- Primary category
- reasoning
- Score direction
- higher
- LLMBoard eligible
- No
- Evaluation key
- browsecomp-long-128k|llm-stats-current
Source-native results are preserved. Eligibility for the overall LLMBoard score is a separate policy decision.
FAQ
Common questions about BrowseComp Long Context 128k.
Which model scores highest on BrowseComp Long Context 128k?
GPT-5.2 is currently ranked first with 92.0%.
What does BrowseComp Long Context 128k measure?
A challenging benchmark for evaluating web browsing agents' ability to persistently navigate the internet and find hard-to-locate, entangled information. Comprises 1,266 questions requiring strategic reasoning, creative search, and interpretation of retrieved content, with short and easily verifiable answers.
Is a higher score better?
Yes. Higher values rank better for this benchmark.
How many models are compared?
5 unique published model results are currently shown.
Does this benchmark affect the overall score?
This benchmark is preserved as source-native evidence but is not eligible for the current overall score.