Image Recognition with LLaVa in Python

[@yuvrajkukreja9727]

Awesome, man! I was not aware of customizing Ollama with this kind of Python script! Thanks :)

[@ammadkhan4687]

Hi, I love all your videos. Could please make a video on getting structured output using ollama. I have use-case to extract specific information from the image and get the output so that automatically the data will be added in database. thanks in advance.

[@NiceTechViews5403]

impressive this llva! my original plan was to detect objects via yolo7 ..give the detected objects to ollama to get some text..and let this text then sound via a loudspeaker. llva ist detecting much more object i guess!? - thx for your video 🙂

[@blackstonesoftware7074]

This is quite useful! It gives me some great ideas for my own local apps!

[@joebywan]

Rad video, thanks dude. Why's the image path take a list, but supplying multiple images to it doesn't work?

[@wasgeht2409]

Thanks :) Is it possible to use this model as an ocr alternativ to get for example informationen from a jpeg image which is an id-card ?

[@sumukhas5418]

This will be too much heavy for just that Instead considering yolo would be a better option

[@wasgeht2409]

@@sumukhas5418 Thanks for the answer :) Actually I am trying pytesseract to read id-card information, which are photographed by a phone and the results are not very good :/ Do you have some ideas, how I could get some better results?

[@derekchance8197]

Are there models that recognize a photo and then vectorizes it?

[@AlissonSantos-qw6db]

Nice, very helpful! Is it possible to create embeddings of pictures with the model?

[@declan6052]

How can I modify this code to use my local GPU? It seems to default to my CPU but can't find any way to do this easily

[@NiceTechViews5403]

it is using my GPU..i have py39, CUDA 11.2 and cuDNN 8, 2019 Visual Studio, GTX 1660TI “Tuning sm_75”

[@R8R809]

Thanks for the video, how to make sure that I install Ollama on the GPU not on the CPU?

[@GuillermoGarcia75]

Riding the awesomeness wave again!

[@rajm5349]

can we get the answer in different languages as per the client requrement just like in hindi or tamil or japanese etc if possible

[@yuvrajkukreja9727]

how to add long term memory in this local llm ???

[@jaykrown]

This was very helpful, my first time getting results from a multimodal LLM directly using Python.

[@brpatil_007]

Is ollama and llava is free to use and I have spec 16GB/1TB RTX 3050Ti what no. of model is suitable for my device 13B one or else. And I already using ollama basic 4GB model in my device is it ok to run 13B model and some Other model like OpenAi or Gemini API??

[@giovannicordova4803]

If my local ram is 8 gb, which ollama model would you recommend to use?

[@WebWizard977]

deepseek-coder ❤

[@WebWizard977]

deepseek-coder ❤

[@aaronbornmann9835]

Thanks for your help you legend

[@timstevens3361]

what gpu ? how much vram ?

[@fastmamajama]

wow this is too easy to be real. i am using opencv to record videos of flying saucers. i could record images and use llama to verify if there is a flying saucer in it. can i also search videos with videos: instead of images:?

[@potatoes1000]

is this fully offline? I am not sure you downloaded the 13B 7.4Gb package

[@naturexmusic2567]

Help me out ,it took less than 10 seconds to get the output , but for me it is like taking 3mins to run , of course it runs , i am happy but it is too late

[@santhosh-j7e]

My computer takes more than an hour , the system is installed with a 4GB 3060 GPU , what can I do

[@naturexmusic2567]

@@santhosh-j7e I dont know man , i was like working it for my hackathon , i tried like all pc ,like pentium , i3 , i5 ,i7 but no difference.

[@Isusgsue]

What a nice vid. Can I do a ai without using open ai ?

[@antonpictures]

rag - webcam - selfawareness - speech --> tutorial pls

[@aoa1015]

How much RAM and VRAM needed ?!

[@RedFoxRicky]

With 4-bit quantization, for LLaVA-1.5-7B, it uses less than 8GB VRAM on a single GPU, typically the 7B model can run with a GPU with less than 24GB memory, and the 13B model requires ~32 GB memory. You can use multiple 24-GB GPUs to run 13B model

[@Justwil07]

7.5 Gb ?????

[@Tech_Distro]

It's 4.7gb for 7b version

[@syedmokarromhossain4867]

First comment 😊😊😊

[@arjuntt2604]

oh im too fast