Our new X account is live! Follow @wizwand_team for updates
Home
/
Benchmarks
Text Generation on MiniGPT-4
Loading...
48.1
BLEU-1
SARE
43.628
44.789
45.95
47.111
Jan 23, 2026
BLEU-1
BLEU-2
BLEU-4
Information Content
Perplexity (PPL)
Updated 4d ago
Evaluation Results
Method
Method
Links
BLEU-1
BLEU-2
BLEU-4
Information Content
Perplexity (PPL)
SARE
Model=MiniGPT-4
2026.01
48.1
32.8
17.2
85.1
0.11
EFUF
Model=MiniGPT-4
2026.01
45.6
31.1
16.7
87.5
0.121
Original
Model=MiniGPT-4
2026.01
43.8
29.5
15.5
86.7
0.134
Feedback
Search any
task
Search any
task
