Skip to content

Benchmark

WideSearch

Scores are percentages. Higher is better.

Models measured
23
Scores recorded
61
Sources
3
IQ weight
0.00

Score distribution

36.3%
Mean
17.3%
Median
29.7%
Std dev
8.5%
10th percentile
79%
90th percentile
4%
Min
4%83.9%
The band covers the middle 80% of scores. The line marks the median.

This benchmark has a weight of 0.00 for chat models, measured over 23 models. The weight is how much it counts toward llminfo IQ. We measure it from the data, not from the raw scores.

Leaderboard

#ModelScore
1Hy4 preview· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim83.9%
2Hy4 previewofficial model-card appendix· Reported by Tencent Hy· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it.83.9%
3Qwen3.8 Max· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim81.9%
4Qwen3.8 MaxQwen表:友商=Claude Code,Qwen=Qwen-Agent · 4次平均 item-F1· Reported by Qwen· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it.81.9%
5Claude Fable 5Qwen表:友商=Claude Code,Qwen=Qwen-Agent · 4次平均 item-F1· Reported by Qwen· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it.81.2%
6Kimi K2.6· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim80.8%
7Kimi K2.5· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim79%
8Hy3 (Deep Infra)· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim76.4%
9Hy3 (Deep Infra)Hy3 official Benchmark Appendix · highest reasoning tier unless noted· Reported by Tencent Hunyuan· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it.76.4%
10Qwen3.7 MaxQwen表:友商=Claude Code,Qwen=Qwen-Agent · 4次平均 item-F1· Reported by Qwen· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it.75.2%
11Qwen3.6 Plus· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim74.3%
12Qwen3.6 PlusQwen official release table · thinking mode unless benchmark requires otherwise · 256K · context management· Reported by Alibaba Qwen· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it.74.3%
13Qwen3.5 397B A17B· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim74%
14Qwen3.5 397B A17BQwen3.5 release/model card · 256K · no context management· Reported by Qwen· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it.74%
15Ling 3.0 Flash· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim73.6%
16Claude Opus 4.8Qwen表:友商=Claude Code,Qwen=Qwen-Agent · 4次平均 item-F1· Reported by Qwen· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it.72.9%
17MiniMax M2.5search/tool figure· Reported by MiniMax· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it.70.3%
18Doubao Seed 1.8· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim63.8%
19Qwen3.5 27B· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim61.1%
20Qwen3.5 122B-A10B· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim60.5%
21Qwen3.6 35B A3B· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim60.1%
22Qwen3.6 FlashQwen3.6-35B-A3B official model card · hosted API name Qwen3.6-Flash· Reported by Alibaba Qwen· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it.60.1%
23Qwen3 Max· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim57.9%
24Qwen3.5 35B A3B· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim57.1%
25GPT-5.6 LunaOpenRouter harness, openai/gpt-5.6-luna, perplexity, server-tool, 25, xhigh· OpenRouter benchmark exports22%
26GPT-5.6 SolOpenRouter harness, openai/gpt-5.6-sol, perplexity, server-tool, 25, high· OpenRouter benchmark exports21%
27Claude Opus 5OpenRouter harness, anthropic/claude-opus-5, perplexity, server-tool, 25, high· OpenRouter benchmark exports20.2%
28DeepSeek V4 Flash 0731OpenRouter harness, deepseek/deepseek-v4-flash-0731, perplexity, server-tool, 25, high· OpenRouter benchmark exports20%
29GPT-5.6 LunaOpenRouter harness, openai/gpt-5.6-luna, perplexity, server-tool, 5, xhigh· OpenRouter benchmark exports18%
30GPT-5.6 SolOpenRouter harness, openai/gpt-5.6-sol, perplexity, server-tool, 5, high· OpenRouter benchmark exports18%
31DeepSeek V4 Flash 0731OpenRouter harness, deepseek/deepseek-v4-flash-0731, perplexity, server-tool, 5, high· OpenRouter benchmark exports17.3%
32GPT-5.6 SolOpenRouter harness, openai/gpt-5.6-sol, native, server-tool, 5, high· OpenRouter benchmark exports17%
33GPT-5.6 SolOpenRouter harness, openai/gpt-5.6-sol, parallel, server-tool, 25, high· OpenRouter benchmark exports16.5%
34GPT-5.6 LunaOpenRouter harness, openai/gpt-5.6-luna, exa, server-tool, 25, xhigh· OpenRouter benchmark exports16%
35GPT-5.6 SolOpenRouter harness, openai/gpt-5.6-sol, native, server-tool, 25, high· OpenRouter benchmark exports16%
36DeepSeek V4 Flash 0731OpenRouter harness, deepseek/deepseek-v4-flash-0731, exa, server-tool, 25, high· OpenRouter benchmark exports15%
37Claude Opus 5OpenRouter harness, anthropic/claude-opus-5, parallel, server-tool, 25, high· OpenRouter benchmark exports13.6%
38GPT-5.6 SolOpenRouter harness, openai/gpt-5.6-sol, parallel, server-tool, 5, high· OpenRouter benchmark exports13.5%
39Claude Opus 5OpenRouter harness, anthropic/claude-opus-5, perplexity, server-tool, 5, high· OpenRouter benchmark exports13.1%
40GPT-5.6 SolOpenRouter harness, openai/gpt-5.6-sol, native, server-tool, 1, high· OpenRouter benchmark exports13%
41GPT-5.6 SolOpenRouter harness, openai/gpt-5.6-sol, parallel, server-tool, 1, high· OpenRouter benchmark exports13%
42GPT-5.6 SolOpenRouter harness, openai/gpt-5.6-sol, perplexity, server-tool, 1, high· OpenRouter benchmark exports13%
43Claude Opus 5OpenRouter harness, anthropic/claude-opus-5, exa, server-tool, 25, high· OpenRouter benchmark exports12.1%
44Claude Opus 5OpenRouter harness, anthropic/claude-opus-5, exa, server-tool, 5, high· OpenRouter benchmark exports12.1%
45GPT-5.6 LunaOpenRouter harness, openai/gpt-5.6-luna, exa, server-tool, 5, xhigh· OpenRouter benchmark exports12%
46GPT-5.6 SolOpenRouter harness, openai/gpt-5.6-sol, exa, server-tool, 25, high· OpenRouter benchmark exports11.6%
47GPT-5.6 SolOpenRouter harness, openai/gpt-5.6-sol, exa, server-tool, 5, high· OpenRouter benchmark exports11.6%
48Claude Opus 5OpenRouter harness, anthropic/claude-opus-5, parallel, server-tool, 5, high· OpenRouter benchmark exports11.5%
49GPT-5.6 LunaOpenRouter harness, openai/gpt-5.6-luna, parallel, server-tool, 25, xhigh· OpenRouter benchmark exports11.5%
50GPT-5.6 LunaOpenRouter harness, openai/gpt-5.6-luna, parallel, server-tool, 5, xhigh· OpenRouter benchmark exports11%

Showing 1–50 of 61 results.

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