Model
string
Text input
float64
ChatIDEFICS
string
Image input
float64
Image input file
float64
Decoding strategy
string
Number of beams
int64
Sampling temperature
float64
N-gram repetition threshold
int64
Maximum number of new tokens to generate
int64
Minimum number of new tokens to generate
int64
Repetition penalty
int64
Length penalty
int64
Top K
int64
Top P
float64
Penalty alpha
float64
flag
float64
username
float64
HuggingFaceM4/idefics-9b-instruct
null
ChatIDEFICS/tmpult4kebk.json
null
null
Top P Sampling
3
1.2
0
512
0
1
1
50
0.8
0.95
null
null