Hello! Just going to ask, can we use LM Studio as back end for vision capable models then connect it to Silly Tavern? I don't understand how to set it in image captioning extensions of Silly Tavern as it's not in the options. Do we need to use an OpenAI-compatible API and set it as Chat Completions instead of Text Completions? Is it possible to use the vision capabilities and support that LM Studio has through API?
I am also facing the same issue and am unable to upload any images in the LM Studio as I don't have an option showing any toggle or any sort to upload images, or to attach images mmproj file which is the vision adapter how to download it , can make an video or please explain it below in comment
@@fahdmirza I am also facing the same issue and am unable to upload any images in the LM Studio as I don't have an option showing any toggle or any sort to upload images, or to attach images
Dear Fahad , is there any way to configure LM studio on SSL,as most of the application of my setups are using htrps, so we neeed to call https an LM studio instead of http
@@fahdmirza sometimes mentioned in the model card itself, else use preset for the base model. If that doesn't work, you can look out for chat template for the model in tokenizer_config file.
Hello! Just going to ask, can we use LM Studio as back end for vision capable models then connect it to Silly Tavern?
I don't understand how to set it in image captioning extensions of Silly Tavern as it's not in the options.
Do we need to use an OpenAI-compatible API and set it as Chat Completions instead of Text Completions? Is it possible to use the vision capabilities and support that LM Studio has through API?
yes you can
@@fahdmirza Okay, thank you! Can you please also elaborate where is the option for this or the steps to make it work?
Great and intuitive explanation
Thank you.
I am also facing the same issue and am unable to upload any images in the LM Studio as I don't have an option showing any toggle or any sort to upload images, or to attach images
mmproj file which is the vision adapter how to download it , can make an video or please explain it below in comment
plz try to upgrade lm studio
i just install LM sutdio and i cant see image uploading tangle next to chat text assistance box ? do you have any idea ?
Please elaborate more
@@fahdmirza I am also facing the same issue and am unable to upload any images in the LM Studio as I don't have an option showing any toggle or any sort to upload images, or to attach images
Dear Fahad , is there any way to configure LM studio on SSL,as most of the application of my setups are using htrps, so we neeed to call https an LM studio instead of http
use nginx for reverse proxy
in lm studio i dont see image icon ... it seems to only take text input for me.
Make sure you have downloaded mmproj file which is the vision adapter
@@fahdmirza where mmproj file download from ?
@@FranchGuy The mproj file of the VISION ADAPTER should be visible in the model directory when you mark it for downloading
Awesome! I love how fast it responded. Which machine type/specs are you using?
I am renting a GPU. Get 50% Discount on any A6000 or A5000 GPU rental, use following link and coupon:
bit.ly/fahd-mirza
Coupon code: FahdMirza
I would suggest you to use proper Preset for the model to give consistent results.
Thanks for great suggestion. What's your way of finding preset for a given model?
@@fahdmirza sometimes mentioned in the model card itself, else use preset for the base model. If that doesn't work, you can look out for chat template for the model in tokenizer_config file.
Can we do RAG on LM studio?
No
This issue is bad, it doesn't recognize new image in the same chat, have to start new chat per image. Hope this gets fixed.
true
Thank you so much brother.
very welcome.
thanks, gentlemen
no problem.
I tried this model. It sucks ballocks lol.
thanks for feedback