The AI arena is free today

Open Superagent

BrowseComp-Plus

PaperImplementation

Progress Over Time

Interactive timeline showing model performance evolution on BrowseComp-Plus

State-of-the-art frontier
Open
Proprietary

BrowseComp-Plus Leaderboard

1 models
ContextCostLicense
1
Inception
Inception
—1.0M$0.10 / $0.35
Notice missing or incorrect data?
About this benchmark

What is BrowseComp-Plus?

BrowseComp-Plus is a fixed-corpus version of BrowseComp for evaluating deep-research agents. It pairs BrowseComp questions with a curated, human-verified document collection so retrieval and agent results are reproducible and comparable across models and retrievers.

BrowseComp-Plus is a text benchmark evaluating models on reasoning, search, and agents tasks. LLM Stats tracks 1 models on this benchmark, scored on a 0–1 scale. The current average is 0.6, with the leader at 0.6.

Compare leaders on the best AI for reasoning, best AI for search and best AI for agents leaderboards.

Current leaders

Mercury 3 from Inception currently leads the BrowseComp-Plus leaderboard with a score of 0.552 across 1 evaluated AI models.

1Mercury 3Inception55.2%

Source paper

Title
BrowseComp-Plus: A More Fair and Transparent Evaluation Benchmark of Deep-Research Agent
Authors
Zijian Chen, Xueguang Ma, Shengyao Zhuang, Ping Nie, and 16 others
Published
Abstract

Deep-Research agents, which integrate large language models (LLMs) with search tools, have shown success in improving the effectiveness of handling complex queries that require iterative search planning and reasoning over search results. Evaluations on current benchmarks like BrowseComp relies on black-box live web search APIs, have notable limitations in (1) fairness: dynamic and opaque web APIs hinder fair comparisons and reproducibility of deep research methods; (2) transparency: lack of control over the document corpus makes it difficult to isolate retriever contributions. In other words, the current evaluations may compare a complete deep research system at a given time, but they do not foster well-controlled experiments to provide insights into the capability of underlying deep research LLMs. To address these challenges, we introduce BrowseComp-Plus, a benchmark derived from BrowseComp, employing a fixed, carefully curated corpus. Each query in BrowseComp-Plus includes human-verified supporting documents and mined challenging negatives, enabling controlled experimentation. The benchmark is shown to be effective in distinguishing the performance of deep research systems. For instance, the open-source model Search-R1, when paired with the BM25 retriever, achieves 3.86% accuracy, whereas the GPT-5 achieves 55.9%. Integrating the GPT-5 with the Qwen3-Embedding-8B retriever further enhances its accuracy to 70.1% with fewer search calls. This benchmark allows comprehensive evaluation and disentangled analysis of deep research agents and retrieval methods, fostering insights into retrieval effectiveness, citation accuracy, and context engineering in Deep-Research system.

FAQ

Common questions about the BrowseComp-Plus benchmark and leaderboard.

What is the BrowseComp-Plus benchmark?

BrowseComp-Plus is a fixed-corpus version of BrowseComp for evaluating deep-research agents. It pairs BrowseComp questions with a curated, human-verified document collection so retrieval and agent results are reproducible and comparable across models and retrievers.

What is the BrowseComp-Plus leaderboard?

The BrowseComp-Plus leaderboard ranks 1 AI models based on their performance on this benchmark. Currently, Mercury 3 by Inception leads with a score of 0.552. The average score across all models is 0.552.

What is the highest BrowseComp-Plus score?

The highest BrowseComp-Plus score is 0.552, achieved by Mercury 3 from Inception.

How many models are evaluated on BrowseComp-Plus?

1 models have been evaluated on the BrowseComp-Plus benchmark, with 0 verified results and 1 self-reported results.

Where can I find the BrowseComp-Plus paper?

The BrowseComp-Plus paper is available at https://arxiv.org/abs/2508.06600. The paper details the methodology, dataset construction, and evaluation criteria.

Where can I find the BrowseComp-Plus dataset?

The BrowseComp-Plus dataset is available at https://github.com/texttron/BrowseComp-Plus.

What categories does BrowseComp-Plus cover?

BrowseComp-Plus is categorized under reasoning, search, and agents. The benchmark evaluates text models.

What's the difference between BrowseComp-Plus and BrowseComp?

BrowseComp-Plus is a variant of BrowseComp. See the BrowseComp leaderboard for the broader benchmark and per-model comparison.

Which model offers the best value on BrowseComp-Plus?

Among models scoring within 10% of the leader, Mercury 3 from Inception is the cheapest, at $0.10 per million input tokens with a score of 0.552.

How recent are the BrowseComp-Plus leaderboard results?

The BrowseComp-Plus leaderboard was last updated in November 2026 and currently includes 1 evaluated models.