Skip to content

Benchmark

BrowseComp-Plus

Agentic benchmarks measure how well models carry out tasks that take several steps. Scores are percentages. Higher is better.

Models measured
19
Scores recorded
24
Sources
3
IQ weight
1.07

Score distribution

59.6%
Mean
65.9%
Median
22.2%
Std dev
29.5%
10th percentile
90.5%
90th percentile
10.7%
Min
10.7%95.2%
The band covers the middle 80% of scores. The line marks the median.

This benchmark has a weight of 1.07 for chat models, measured over 19 models. The weight is how much it counts toward llminfo IQ. We measure it from the data, not from the raw scores.

Leaderboard

#ModelScore
1GPT 5· BrowseComp-Plus data95.2%
2GLM 5.1· BrowseComp-Plus data90.7%
3GPT 5.5· BrowseComp-Plus data90.5%
4Claude Opus 4.5launch-page footnote · full context-management configuration· Reported by Anthropic· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it.85.3%
5GPT-5 mini· BrowseComp-Plus data80%
6Claude Opus 4.5Qwen3-Embedding-8B index · tool-result clearing + memory/context reset · Sonnet 4.5 grader· Reported by Anthropic· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it.72.9%
7GPT-6 Luna· BrowseComp-Plus data71.9%
8Claude Opus 4.5launch-page footnote · basic fetch configuration· Reported by Anthropic· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it.70.5%
9Tongyi DeepResearch 30B A3B· BrowseComp-Plus data68.1%
10Claude Opus 4.5Qwen3-Embedding-8B index · tool-result clearing · no fetch · Sonnet 4.5 grader· Reported by Anthropic· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it.67.6%
11Claude Sonnet 4.5Qwen3-Embedding-8B index · tool-result clearing + memory/context reset · Sonnet 4.5 grader· Reported by Anthropic· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it.67.2%
12OpenAI o3· BrowseComp-Plus data65.9%
13GLM 4.7· BrowseComp-Plus data64.7%
14Claude Sonnet 4.5Qwen3-Embedding-8B index · tool-result clearing · no fetch · Sonnet 4.5 grader· Reported by Anthropic· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it.60.4%
15Mercury 3· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim55.2%
16Claude Haiku 4.5Qwen3-Embedding-8B index · tool-result clearing + memory/context reset · Sonnet 4.5 grader· Reported by Anthropic· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it.54.7%
17Claude Haiku 4.5Qwen3-Embedding-8B index · tool-result clearing · no fetch · Sonnet 4.5 grader· Reported by Anthropic· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it.52.5%
18GLM 4.6· BrowseComp-Plus data48.2%
19Gemini 2.5 Flash· BrowseComp-Plus data39.9%
20GPT 4.1· BrowseComp-Plus data36.4%
21kimi-k2-0711-preview· BrowseComp-Plus data35.4%
22Gemini 2.5 Pro· BrowseComp-Plus data29.5%
23DeepSeek R1· BrowseComp-Plus data16.4%
24Qwen3 32B· BrowseComp-Plus data10.7%

Data credits: BenchBoard by SYuan03 (github.com/SYuan03/benchboard). MIT covers its code and data organization; no results licence is inferred. Numeric relays retain the original stating-source links. · Data by LLM Stats