moondream/moondream3-preview
Image-Text-to-Text • 9B • Updated • 7.59k • 601
AI & ML interests
None defined yet.
Recent Activity
View all activity
Call post_init() in HfMoondream for Transformers 5 compatibility
1
#33 opened 23 days ago
by 
err805
Call post_init() in HfMoondream for Transformers 5 compatibility
1
#33 opened 23 days ago
by err805
err805
updated a
model 23 days ago
AttributeError: 'HfMoondream' object has no attribute 'all_tied_weights_keys'. Did you mean: '_tied_weights_keys'?
8
#30 opened about 2 months ago
by 
Clausss
err805
updated a
model 26 days ago
Update BF16 weights + code to modelv2 shards (region LN + finetune support)
#32 opened about 2 months ago
by err805
Update BF16 weights + code to modelv2 shards (region LN + finetune support)
#32 opened about 2 months ago
by err805
vikhyatk
updated a
model 2 months ago
err805
updated a
dataset 3 months ago
err805
published a
dataset 3 months ago
vikhyatk
published a
dataset 3 months ago
vikhyatk
updated a
dataset 3 months ago
vikhyatk
published a
model 3 months ago
err805
updated a
dataset 3 months ago
err805
published a
dataset 3 months ago
err805
updated a
dataset 3 months ago
Post
4708
Announcing RefCOCO-M, a refreshed RefCOCO with pixel-accurate masks and the problematic prompts removed.
moondream/refcoco-m
moondream/refcoco-m
Post
5720
Just released a preview of Moondream 3! moondream/moondream3-preview
This is a 9B parameter, 2B active MoE VLM with state of the art visual reasoning capabilities.
More details in the release blog post: https://moondream.ai/blog/moondream-3-preview
This is a 9B parameter, 2B active MoE VLM with state of the art visual reasoning capabilities.
More details in the release blog post: https://moondream.ai/blog/moondream-3-preview
- 3 replies
Post
6225
🚨 New VQA + captioning dataset! moondream/megalith-mdqa
Images from Megalith, captioned using Moondream, then transformed to short-form QA.
9M+ images, 6-10 QA pairs per image.
Images from Megalith, captioned using Moondream, then transformed to short-form QA.
9M+ images, 6-10 QA pairs per image.
Post
6951
Just released a dataset with 7000+ hours of synthetically generated lo-fi music. vikhyatk/lofi