Wikibench

Seal-0

reasoning

Seal-0 is a benchmark for evaluating agentic search capabilities, testing models' ability to navigate and retrieve information using tools.

Leaderboard

Showing 6 of 6 results

  1. Kimi K2.5
    57.4%
    i
  2. Kimi K2-Thinking-0905
    56.3%
    i
  3. Qwen3.5-27B
    47.2%
    i
  4. Qwen3.5-397B-A17B
    46.9%
    i
  5. Qwen3.5-122B-A10B
    44.1%
    i
  6. Qwen3.5-35B-A3B
    41.4%
    i