Grok Vision - First Multimodal Model from XAi

X.ai just made an announcement about Grok-1.5 Vision. Its their new multimodal model that can understand images and can write code based on flow diagrams just like GPT-4 :)
🦾 Discord: / discord
☕ Buy me a Coffee: ko-fi.com/promptengineering
|🔴 Patreon: / promptengineering
💼Consulting: calendly.com/engineerprompt/c...
📧 Business Contact: [e-mail]
Become Member: tinyurl.com/y5h28s6h
💻 Pre-configured localGPT VM: bit.ly/localGPT (use Code: PromptEngineering for 50% off).
Signup for Advanced RAG:
tally.so/r/3y9bb0
LINKS:
x.ai/blog/grok-1.5v
All Interesting Videos:
Everything LangChain: • LangChain
Everything LLM: • Large Language Models
Everything Midjourney: • MidJourney Tutorials
AI Image Generation: • AI Image Generation Tu...

[@ilianos]

Can you test more diagrams? For example, I tried several vision models (including ChatGPT, Moondream2, Llava-1.6) to analyse venn diagrams (on AI or fine-tuning) but most of them failed with recognizing shapes/labels/colors in the details (e.g. of overlapping labels and colors). So for me, they're all still far from perfect.

[@Artorias920]

wouldnt it be hilarious if OpenAI loses their entire moat

[@engineerprompt]

I agree, seems like other companies are catching up fast.

[@disco4535]

Its a matter of when

[@venkyvings934]

Hi bro I recently subscribed to your channel for prompt engineering.....but I'm confused about where to start........

[@engineerprompt]

I think this will be a good video to get started :) kzbin.info/www/bejne/oJC6ep9vob-ngLc

[@venkyvings934]

@@engineerprompt thank you

[@ps3301]

How big is the model ? 300b?

[@engineerprompt]

Not sure about the exact size

[@rabeemohammed5351]

I have found several explanations for you explaining how to create a local conversation with files and without the Internet, but I want you to give me the best model that you deem appropriate for conversation with files without the Internet. Thank you very much for the wonderful content that you provide to us ❤❤❤.

[@engineerprompt]

I would say the best model that you can run for that purpose is Mistral-7B or the bigger sibling 8x7B.

[@rabeemohammed5351]

@@engineerprompt I ask you to help us make a video on how to use it locally and how it can be linked to the files that we want to have a conversation on. I ask you to respond. Will a video be made on this idea or provide a link to the videos that may help me do this? Thank you and your efforts ❤️

[@simonpeter9617]

Not good enough, waiting for gpt5

[@helix8847]

haha, what a dumb statement. GPT5 aint going to blow us away like you think it will kid. They just updated their GPT 4 and its nothing special again. Claude 3 craps all over GPT 4. Stop licking OpenAI and Sams balls kid. You should be happy that we have good competition because if OpenAI did not have comp you would not be able to afford to use it. Their free version is garbage.

[@HistoryIsAbsurd]

Yeah but its XAI so that instantly puts it in the "yeah it probably sucks balls" category. Ill bet 5 channel memberships on it That is...if anyone can actually run the dang thing without 3 A100s.. I dont care abut hype. I care about functionality and cost. This channel is increasingly moving towards the "just reporting on new hype" and less on the "teaching and prompt engineering" .. Its why i unsubbed to Mattvidpro and Matther Bermann.