Abacus ChatLLM - Great alternative to ChatGPT + RAG

Discover the incredible capabilities of Abacus AI's Chat LLM interface! 🌐 In this video, we delve into how to create and deploy custom chatbots using the powerful features of Abacus AI. From integrating multiple language models to fine-tuning and RAG (Retrieval-Augmented Generation), we cover it all. 🚀
Sign Up for Abacus AI (First Month Free) bit.ly/abacusa...
📝 What You Will Learn:
Interface Overview: Get familiar with the Chat LLM interface.
Data Upload & Management: Learn to upload PDFs and other documents.
RAG Setup: Understand how to set up document retrieval.
Fine-Tuning: Discover how to fine-tune your language model for better accuracy.
Deployment: Step-by-step guide to deploying your chatbot.
Slack Integration: See how to integrate your chatbot with Slack for seamless communication.
📌 Key Features:
Multiple Language Models: Choose from a variety of models for your specific use case.
Cost-Effective: Pay less than $10 per user, with the first month free!
Custom Chatbots: Leverage both fine-tuning and RAG for optimal performance.
Team Collaboration: Easily invite team members and manage user permissions.
Integration: Connect with multiple services and platforms effortlessly.
🔗 Links:
Patreon: / mervinpraison
Ko-fi: ko-fi.com/merv...
Discord: / discord
Twitter / X : / mervinpraison
Sponsor a Video or Do a Demo of Your Product: mer.vin/contact/
🔖 Timestamps:
0:00 - Introduction to Abacus AI Chat LLM
0:15 - Key features and benefits
1:00 - Combining fine-tuning and RAG techniques
1:10 - Steps to create a custom LLM chatbot
1:40 - Document management and embedding process
3:00 - Setting up and deploying the chat interface
5:30 - Uploading data and creating datasets
7:04 - Training and evaluating the model
9:00 - Using the custom-built chat model
9:41 - Integrating the chatbot with Slack
🌟 Don't Forget to Like, Share, and Subscribe!
If you found this video helpful, make sure to give it a thumbs up, share it with your network, and subscribe for more AI tutorials. Hit the bell icon to stay updated with our latest content! 🔔

[@tuaitituaiti1565]

Thank you again watching your content then implementation of the learning is rewarding. Thank again.

[@marki17x]

Beware!!! Abacus asking credit cards but NO possibllity cancel order

[@RafaelFleury_00]

I found their site a month ago. You have to send a email request to their team to delete the acc. My acc will be deleted this month day 17 along with the cancellation of the plan.

[@aimattant]

They have a cancel order feature in profile section

[@vikinist]

@@aimattant for me its asking for cred details after payment also

[@micbab-vg2mu]

At the moment I use Poe to test output quality of diffrent models - I will try Abacus :)

[@burak_gulmez]

Can i train llm using lots of pdf files? I do not have questions and answers file. Just using pdf files? Is it possible on abacus ai?

[@farexBaby-ur8ns]

You say the data is kept private. But i am uploading office docs to yonder for rag-ing and finetuning - or did i get that wrong?
Are there alternatives to abacus ai?
Thx for this vid. Never knew that there existed something like this. Very appreciative of all yr vids- farthest mileage in 10- 15mins(the outer edge of my attn span these days😅)

[@ssatish811]

when I try to create dataset, I don't see the evaluation option in my screen . Any reasons ? @5.55 seconds

[@aisprint]

Thank you for the introduction video of Abacus. It’s amazing that we can train models with no code. If you know, please answer the following questions.
Is fine-tuning possible with any model? From my understanding, GPT, Claude, and Gemini are closed models, so I thought they couldn’t be fine-tuned from the outside (excluding gpt3.5turbo, gpt4turbo fine tune models).
Is the fine-tuning being performed here more like saving the uploaded Q&A as additional documents, similar to RAG (so with no changes to the LLM itself), rather than actually retraining the model?

[@MervinPraison]

Fine-tuning not for any models. There is a list of models Abacus AI offer.
It is actual retraining the model.
Both RAG and Fine Tuned model will work together to get us the final result.

[@aisprint]

Thank you, Marvin!

[@maxopbgmigamer]

Please post video for build custom chatbot with chat data like ghostbot

[@behroozkarjoo6092]

can you add websearch to the pipeline?

[@orthodox_gentleman]

Yes

[@vitalis]

what's an open source alternative?

[@tonywhite4476]

I hope they're sposoring your video otherwise it's promo pro bono.

[@john_blues]

It's likely sponsored, plus he has their affiliate link in the description.

[@Alex29196]

Definitely not willing to pay for experimentation.