Wikibench

V*

reasoning

Multilingual: no

A visual reasoning benchmark evaluating multimodal inference under challenging spatial and grounded tasks.

Methodology

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

Leaderboard

Showing 7 of 7 results

  1. Kimi K2.6 self-reported llm-stats
    96.9%
  2. Qwen3.6 Plus self-reported llm-stats
    96.9%
  3. Qwen3.6-27B self-reported llm-stats
    94.7%
  4. Qwen3.5-27B self-reported llm-stats
    93.7%
  5. Qwen3.5-122B-A10B self-reported llm-stats
    93.2%
  6. Qwen3.5-35B-A3B self-reported llm-stats
    92.7%
  7. GLM-5V-Turbo self-reported llm-stats
    89.0%