Remove In-place operations
#40 opened 17 days ago
by
yyyyifan
Increasing context length for language input
1
#39 opened about 1 month ago
by
chrishoertnagl

Add HF Chat template
1
#38 opened about 2 months ago
by
anvdn
Multi-turn Chat History Management
2
#37 opened 3 months ago
by
Jaykumaran17
Do you support multi-turn conversations?
4
#36 opened 3 months ago
by
whopeople
Beam search functionality?
1
#35 opened 3 months ago
by
adhiraj1998
Multi-image input support
4
#34 opened 3 months ago
by
cynricfu
How should I extract attention maps? Can you provide a specific example?
5
#33 opened 3 months ago
by
whopeople
force json output
1
#32 opened 3 months ago
by
joa23
How to finetune using DPO?
7
#31 opened 3 months ago
by
Maverick17
Create handler.py
2
#29 opened 4 months ago
by
shrijayan

Any plans on when vllm will be supported?
8
#26 opened 4 months ago
by
karlyukang
Image size versus inference speed/accuracy
3
#22 opened 5 months ago
by
logankeenan
Apple Silicon not running with device type mps and acceleration
6
#21 opened 5 months ago
by
Troubadix
Use Molmo vision encoder for classification.
#20 opened 5 months ago
by
shafeeq007
how to use the annotation mode
1
#15 opened 5 months ago
by
mehmetsat

Finetuning script for Molmo
#14 opened 5 months ago
by
2U1
batch inference supported?
6
#7 opened 5 months ago
by
chenkq
Could you let me know when the bfloat16 model will be uploaded? I can't run the float32 model!
7
#5 opened 5 months ago
by
Cach
Molmo-7B-D-0924 OOM on A100 80GB using Quick Start code
5
#1 opened 5 months ago
by
sasawq21