Share your thoughts, 1 month free Claude Pro on us
See more
Home
/
Benchmarks
Scientific Graphics Program Synthesis on DaTikZ v3 (test)
Loading...
88.29
DSim
SciTikZer-8B
71.8788
76.1394
80.4
84.6606
Apr 7, 2026
DSim
KID
cBLEU
TED
Compile Rate
Updated 3mo ago
Evaluation Results
Method
Method
Links
DSim
KID
cBLEU
TED
Compile Rate
SciTikZer-8B
Model size=8B
2026.04
88.29
1.14
16.17
48.83
94.46
Gemini-2.5-Pro
2026.04
87.34
2.86
12.29
50.45
66.42
DeTikZify-v2.5-8b
Model size=8b
2026.04
85.05
1.72
12.83
51.1
90.41
Qwen3-VL-235B-A22B
Model size=235B-A22B
2026.04
83.91
2.66
16.05
49.8
87.82
Qwen3-VL-8B
Model size=8B
2026.04
78.25
5.32
7.75
60.39
55.17
InternVL3.5-14B
Model size=14B
2026.04
72.51
7.7
10.53
53.45
65.31
Feedback
Search any
task
Search any
task