Skip to content

Benchmark

BrowseComp

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

Models measured
84
Scores recorded
203
Sources
4
IQ weight
2.43

Score distribution

66.4%
Mean
74.7%
Median
21.8%
Std dev
35.2%
10th percentile
90.1%
90th percentile
0.6%
Min
0.6%92.2%
The band covers the middle 80% of scores. The line marks the median.

This benchmark has a weight of 2.43 for chat models, measured over 84 models. The weight is how much it counts toward llminfo IQ. We measure it from the data, not from the raw scores.

Leaderboard

#ModelScore
151Qwen3 Max· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim53.9%
152DeepSeek V4 Flash· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim53.5%
153DeepSeek V4 FlashDeepSeek V4 official model card · Think High· Reported by DeepSeek· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it.53.5%
154DeepSeek V3.2Thinking mode · temperature 1.0 · 128K context · commercial search API · baseline · Section 4.4· Reported by DeepSeek· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it.53.4%
155Claude Opus 5OpenRouter harness, anthropic/claude-opus-5, exa, server-tool, 5, high· OpenRouter benchmark exports52.5%
156Step 3.5 FlashStep 3.5 Flash official headline table · vanilla· Reported by StepFun· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it.51.6%
157o4 Mini· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim
Values from 2 sources
51.5%
158DeepSeek Reasoner· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim51.4%
159DeepSeek V3.2· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim
Values from 2 sources
51.4%
160DeepSeek V3.2Thinking mode · temperature 1.0 · 128K context · commercial search API · without context management · Table 2· Reported by DeepSeek· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it.51.4%
161OpenAI o3· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim
Values from 2 sources
49.7%
162Sarvam 105B· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim
Values from 2 sources
49.5%
163Step 3.5 FlashStep 3.5 technical report · Table 17 · 200-item subset · plain context · 86 real steps· Reported by StepFun· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it.49.5%
164Solar Pro 4· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim49.2%
165Mistral Medium 3.5· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim48.6%
166GPT-5.6 LunaOpenRouter harness, openai/gpt-5.6-luna, exa, server-tool, 5, xhigh· OpenRouter benchmark exports48%
167GPT-5.6 SolOpenRouter harness, openai/gpt-5.6-sol, exa, server-tool, 1, high· OpenRouter benchmark exports46.7%
168GPT-5.6 SolOpenRouter harness, openai/gpt-5.6-sol, perplexity, server-tool, 1, high· OpenRouter benchmark exports46.3%
169MiMo V2 FlashMiMo-V2-Flash final evaluation · standard· Reported by Xiaomi MiMo· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it.45.4%
170GLM 4.6· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim
Values from 2 sources
45.1%
171Grok 4 Fast· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim
Values from 2 sources
44.9%
172Nemotron 3 Ultra· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim44.4%
173Nemotron 3 UltraNeMo Evaluator SDK · official table· Reported by NVIDIA· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it.44.4%
174MiniMax M2· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it.
Values from 2 sources
44%
175GPT-5.6 SolOpenRouter harness, openai/gpt-5.6-sol, parallel, server-tool, 1, high· OpenRouter benchmark exports43.7%
176GLM 4.7 Flash· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim
Values from 2 sources
42.8%
177GPT-5.6 LunaOpenRouter harness, openai/gpt-5.6-luna, parallel, server-tool, 5, xhigh· OpenRouter benchmark exports42.5%
178GPT-5.6 SolOpenRouter harness, openai/gpt-5.6-sol, native, server-tool, 1, high· OpenRouter benchmark exports41.8%
179DeepSeek V3.2 Exp· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim
Values from 2 sources
40.1%
180Nemotron 3.5 Lightning 30B A3B· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim37%
181Nemotron 3.5 Lightning 30B A3Bofficial release table· Reported by NVIDIA· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it.37%
182Claude Opus 5OpenRouter harness, anthropic/claude-opus-5, perplexity, server-tool, 1, high· OpenRouter benchmark exports35.8%
183Sarvam-30B· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim
Values from 2 sources
35.5%
184DeepSeek V4 Flash 0731OpenRouter harness, deepseek/deepseek-v4-flash-0731, parallel, server-tool, 5, high· OpenRouter benchmark exports35.2%
185GPT-5.6 LunaOpenRouter harness, openai/gpt-5.6-luna, perplexity, server-tool, 1, xhigh· OpenRouter benchmark exports33.7%
186GPT-5.6 LunaOpenRouter harness, openai/gpt-5.6-luna, exa, server-tool, 1, xhigh· OpenRouter benchmark exports33%
187Nemotron 3 Super 120B A12B· Steel.dev Agent Leaderboards data
Values from 2 sources
31.3%
188Nemotron 3 Super 120B A12BNeMo Evaluator / mostly NeMo Skills · with Search · internal SerpAPI implementation· Reported by NVIDIA· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it.31.3%
189Claude Opus 5OpenRouter harness, anthropic/claude-opus-5, parallel, server-tool, 1, high· OpenRouter benchmark exports30.8%
190DeepSeek V3.1· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim
Values from 2 sources
30%
191Claude Opus 5OpenRouter harness, anthropic/claude-opus-5, exa, server-tool, 1, high· OpenRouter benchmark exports29.4%
192GLM 4.5· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim
Values from 2 sources
26.4%
193GPT-5.6 LunaOpenRouter harness, openai/gpt-5.6-luna, parallel, server-tool, 1, xhigh· OpenRouter benchmark exports23.7%
194GLM 4.5 Air· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim
Values from 2 sources
21.3%
195DeepSeek V4 Flash 0731OpenRouter harness, deepseek/deepseek-v4-flash-0731, parallel, server-tool, 1, high· OpenRouter benchmark exports20%
196Claude Opus 5OpenRouter harness, anthropic/claude-opus-5, exa, plugin, medium· OpenRouter benchmark exports12%
197Claude Opus 5OpenRouter harness, anthropic/claude-opus-5, perplexity, plugin, medium· OpenRouter benchmark exports11%
198o1· Steel.dev Agent Leaderboards data9.9%
199DeepSeek R1· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim
Values from 2 sources
8.9%
200Sonar Deep Research· Steel.dev Agent Leaderboards data6%

Showing 151–200 of 203 results.

Data credits: Steel.dev Agent Leaderboards, MIT (licence) · Data by LLM Stats · 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. · OpenRouter public benchmark leaderboard and search-lane exports, CC BY 4.0. Fractional source scores converted to percentages; immutable dated snapshots. (licence)