Wikibench

MMBench

reasoning official site →

Multilingual: no

A bilingual benchmark for assessing multi-modal capabilities of vision-language models through multiple-choice questions in both English and Chinese, providing systematic evaluation across diverse vision-language tasks with robust metrics.

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 8 of 8 results

  1. Qwen2.5 VL 72B Instruct self-reported llm-stats
    88.0%
  2. Phi-4-multimodal-instruct self-reported llm-stats
    86.7%
  3. Qwen2-VL-72B-Instruct self-reported llm-stats
    86.5%
  4. Qwen2.5 VL 7B Instruct self-reported llm-stats
    84.3%
  5. Phi-3.5-vision-instruct self-reported llm-stats
    81.9%
  6. DeepSeek VL2 Small self-reported llm-stats
    80.3%
  7. DeepSeek VL2 self-reported llm-stats
    79.6%
  8. DeepSeek VL2 Tiny self-reported llm-stats
    69.2%