-
Partly issue, partly solution. The text_model.py for moondream doesn't seem to take advantage of the gpu. Changing device_map to auto accelerated the inference speed on my 3090 from 20s to .5s
`sel…
-
-
While using gradio_demo.py, I noticed that anytime time a response starts with "No" it seems to drop the first character. For example : "o, UPS does not offer pickup services for returns at home. The…
-
![image](https://github.com/Hangover3832/ComfyUI-Hangover-Moondream/assets/57288401/a13473a3-6d04-40ee-9724-bd3e307c891e)
-
Is there a way to compare the embedding of the image to embedding of the text and calculate similarity score?
-
**Describe the bug**
GPTSampler and GPTLoaderSimple stopped working after update.
**To Reproduce**
Just Start
**Screenshots**
**Desktop (please complete the following information):**
…
Aljnk updated
6 months ago
-
Thanks for this work, it was working great on my local machine until today. I don't know what I changed*, I only get a string of exclamation signs as response to any question that I ask. Any hints?
…
-
It does not understand questions like where is this image taken or what meal is this
-
Hi,
running sample.py fails, quick fix was to pass tokenizer as third param
answer = moondream.answer_question(image_embeds, prompt,tokenizer)
https://github.com/vikhyat/moondream/blob/de750a63ec…
-
It would be very interesting to run a vision model like LLAVA, or the very impressive and tiny new Moondream LLM, in the browser. It would be fantastic if it could analyze images and (live) video.
…