Wikibench

WideSearch

reasoning

Multilingual: no

WideSearch is an agentic search benchmark that evaluates models' ability to perform broad, parallel search operations across multiple sources. It tests wide-coverage information retrieval and synthesis capabilities.

Methodology

Imported from llm-stats public benchmark metadata. Modality: text. Max score: 1. Categories: agents, reasoning, search. Language: en. Verified by llm-stats: no.

Leaderboard

Showing 7 of 7 results

  1. Kimi K2.6 self-reported llm-stats
    80.8%
  2. Kimi K2.5 self-reported llm-stats
    79.0%
  3. Qwen3.6 Plus self-reported llm-stats
    74.3%
  4. Qwen3.5-397B-A17B self-reported llm-stats
    74.0%
  5. Qwen3.5-27B self-reported llm-stats
    61.1%
  6. Qwen3.5-122B-A10B self-reported llm-stats
    60.5%
  7. Qwen3.5-35B-A3B self-reported llm-stats
    57.1%