Wikibench

VLMsAreBlind

reasoning

Multilingual: no

A vision-language benchmark that probes blind spots and brittle reasoning in multimodal models.

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 4 of 4 results

  1. Qwen3.5-35B-A3B self-reported llm-stats
    97.0%
  2. Qwen3.6-27B self-reported llm-stats
    97.0%
  3. Qwen3.5-27B self-reported llm-stats
    96.9%
  4. Qwen3.5-122B-A10B self-reported llm-stats
    96.7%