rerun notebook, added baseline, models, prompt engineering

Author: haesleinhuepf

docs/80_benchmarking_llms/20_vision_models.ipynb

@@ -1,26 +1,26 @@
-,gpt-4-turbo-2024-04-09,gpt-4o-2024-05-13,gpt-4o-2024-08-06,gpt-4o-2024-11-20
-0,93,56,55,67
-1,61,52,55,73
-2,54,62,53,50
-3,49,53,70,72
-4,68,73,57,74
-5,53,58,50,53
-6,54,52,56,67
-7,64,56,68,68
-8,61,47,50,52
-9,51,54,53,61
-10,46,65,63,92
-11,53,58,54,57
-12,63,53,46,58
-13,58,47,52,54
-14,76,66,49,64
-15,62,53,68,72
-16,47,48,54,97
-17,83,57,53,78
-18,54,51,49,64
-19,56,61,54,67
-20,73,62,49,53
-21,49,58,58,80
-22,60,58,57,62
-23,54,48,46,70
-24,64,49,65,68
+,gpt-4-turbo-2024-04-09,gpt-4o-2024-05-13,gpt-4o-2024-08-06,gpt-4o-2024-11-20,gpt-4.1-mini-2025-04-14,gpt-4.1-nano-2025-04-14
+0,52,58,47,58,52,54
+1,49,57,47,49,48,64
+2,53,49,52,54,72,53
+3,68,49,53,57,61,75
+4,63,72,49,67,49,65
+5,56,47,52,54,54,77
+6,63,54,53,52,62,34
+7,68,93,56,56,54,53
+8,49,70,57,64,54,41
+9,53,56,58,63,67,35
+10,61,78,53,68,49,37
+11,57,53,54,51,76,48
+12,46,68,52,66,49,51
+13,67,79,62,72,59,60
+14,51,55,54,53,67,58
+15,53,87,50,78,56,47
+16,93,68,63,62,44,41
+17,52,58,54,72,45,54
+18,52,68,53,74,52,66
+19,49,55,62,64,54,56
+20,52,63,62,58,49,44
+21,57,63,54,57,52,41
+22,51,60,52,65,52,57
+23,72,59,70,48,44,53
+24,42,48,55,63,59,41

docs/80_benchmarking_llms/blob_count_by_llms.csv

@@ -0,0 +1,26 @@
+,unspecific,w/o border,with border
+0,61.0,30.0,48
+1,57.0,47.0,56
+2,52.0,50.0,57
+3,65.0,43.0,58
+4,59.0,31.0,58
+5,52.0,,68
+6,67.0,27.0,61
+7,58.0,26.0,66
+8,60.0,41.0,47
+9,57.0,38.0,54
+10,56.0,31.0,61
+11,54.0,43.0,58
+12,52.0,44.0,49
+13,59.0,,62
+14,57.0,55.0,52
+15,67.0,42.0,57
+16,54.0,38.0,81
+17,55.0,30.0,55
+18,47.0,36.0,46
+19,63.0,39.0,48
+20,41.0,18.0,52
+21,,37.0,58
+22,51.0,43.0,63
+23,52.0,35.0,47
+24,56.0,44.0,48