Stanford UniversityAI Measurement ScienceAIMS — AI Measurement Science

Science

AI2D_TEST

Build AI2D_TEST response matrix from VLMEval/OpenVLMRecords

3,088items
254subjects
CC-BY-SA-4.0license
sciencedomain
knowledgedomain
textmodality
imagemodality
item-level responses released
Saturation status: Unknown
Original sourcePaper← All benchmarks

Response matrix

Fit to width. Hover for subject & item; click a cell for details.

AI2D_TEST response matrix: AI models (rows) against items (columns)
Correct (1)Incorrect (0)Unobserved

Scale: 1 = correct · 0 = incorrect

Subjects

  1. 1HunYuan-Standard-Vision0.9606
  2. 2Taiyi0.9077
  3. 3InternVL2_5-78B-MPO0.8922
  4. 4Qwen2.5-VL-72B-Instruct0.8918
  5. 5InternVL2_5-78B0.8912
  6. 6Step1o0.8909
  7. 7Qwen2-VL-72B-Instruct0.8899
  8. 8Qwen2.5-VL-72B0.8886
  9. 9MUG-U-7B0.8883
  10. 10Qwen-VL-Max-08090.8867
  11. 11TeleMM0.8854
  12. 12Ovis2-34B0.8831
  13. 13DoubaoVL0.8831
  14. 14SenseChat-Vision0.8804
  15. 15InternVL2_5-38B-MPO0.8792
  16. 16GPT4.50.8767
  17. 17InternVL2_5-38B0.8756
  18. 18InternVL2-76B0.8756
  19. 19Step1V0.8753
  20. 20BailingMM-Pro-01200.8724
  21. 21JTVL0.8718
  22. 22InternVL2-40B0.8682
  23. 23Ovis2-8B0.8663
  24. 24Ovis1.6-Gemma2-27B0.8656
  25. 25GLM4V_PLUS_202501110.8639
  26. 26Ovis2-16B0.863
  27. 27llava_onevision_qwen2_72b_ov0.862
  28. 28InternVL2_5-26B0.8617
  29. 29InternVL2_5-26B-MPO0.8617
  30. 30ola0.8611
  31. 31MiniCPM-o-2_60.8608
  32. 32bailingMM-Lite0.8601
  33. 33ChatGPT4o0.8585
  34. 34Ovis2-4B0.8565
  35. 35llava_onevision_qwen2_72b_si0.8548
  36. 36BlueLM_V0.8533
Full data on Hugging FaceBack to the gallery