yonatanbitton commited on
Commit
efbe573
1 Parent(s): c55dc37

Update visitbench_leaderboard_Single~Image_Oct282023.tsv

Browse files
visitbench_leaderboard_Single~Image_Oct282023.tsv CHANGED
@@ -2,16 +2,16 @@ Category Model Elo # Matches Win vs. Reference (w/ # ratings)
2
  Single Image human_verified_reference 1361 6030 ---
3
  Single Image LLaVA-Plus 1206 724 30.15% (n=136)
4
  Single Image LLaVA 13B 1091 5474 18.53% (n=475)
5
- Single Image lynx(7B)_v2 prediction 1078 708 15.15% (n=132)
6
- Single Image mPLUG-Owl prediction 1076 5465 16.04% (n=480)
7
- Single Image LlamaAdapter-v2 prediction 1055 5485 14.14% (n=488)
8
- Single Image idefics9b_prediction 1030 842 9.72% (n=144)
9
- Single Image Lynx(8B) predictions 1012 827 11.43% (n=140)
10
- Single Image instruct_blip_output 995 5505 14.12% (n=503)
11
  Single Image otter 970 5495 7.01% (n=499)
12
- Single Image visual_gpt_davinci003_output 937 5486 1.57% (n=510)
13
- Single Image Octopus V2 prediction 936 820 8.90% (n=146)
14
- Single Image MiniGPT-4 prediction 899 5473 3.36% (n=506)
15
  Single Image openflamingo 831 5490 2.95% (n=509)
16
- Single Image panda_gpt_13b_output 767 5480 2.70% (n=519)
17
- Single Image mmgpt_output 757 5504 0.19% (n=527)
 
2
  Single Image human_verified_reference 1361 6030 ---
3
  Single Image LLaVA-Plus 1206 724 30.15% (n=136)
4
  Single Image LLaVA 13B 1091 5474 18.53% (n=475)
5
+ Single Image Lynx 7B V2 1078 708 15.15% (n=132)
6
+ Single Image mPLUG-Owl 1076 5465 16.04% (n=480)
7
+ Single Image LlamaAdapter-v2 1055 5485 14.14% (n=488)
8
+ Single Image idefics9b 1030 842 9.72% (n=144)
9
+ Single Image Lynx(8B) 1012 827 11.43% (n=140)
10
+ Single Image InstructBLIP 995 5505 14.12% (n=503)
11
  Single Image otter 970 5495 7.01% (n=499)
12
+ Single Image visual_gpt_davinci003 937 5486 1.57% (n=510)
13
+ Single Image Octopus V2 936 820 8.90% (n=146)
14
+ Single Image MiniGPT-4 899 5473 3.36% (n=506)
15
  Single Image openflamingo 831 5490 2.95% (n=509)
16
+ Single Image panda_gpt_13b 767 5480 2.70% (n=519)
17
+ Single Image MMGPT 757 5504 0.19% (n=527)