Wikibench

SimpleVQA

multimodal

Multilingual: no

SimpleVQA is a visual question answering benchmark focused on simple queries.

Methodology

Imported from llm-stats public benchmark metadata. Modality: multimodal. Max score: 100. Categories: general, image_to_text, multimodal, vision. Language: en. Verified by llm-stats: no.

Leaderboard

Showing 9 of 9 results

  1. GLM-5V-Turbo self-reported llm-stats
    78.2%
  2. Muse Spark self-reported llm-stats
    71.3%
  3. Kimi K2.5 self-reported llm-stats
    71.2%
  4. Qwen3.6 Plus self-reported llm-stats
    67.3%
  5. Qwen3.5-122B-A10B self-reported llm-stats
    61.7%
  6. Qwen3 VL 235B A22B Thinking self-reported llm-stats
    61.3%
  7. Qwen3.5-35B-A3B self-reported llm-stats
    58.3%
  8. Qwen3.6-27B self-reported llm-stats
    56.1%
  9. Qwen3.5-27B self-reported llm-stats
    56.0%