Groq’s Post

🌋𝗩𝘂𝗹𝗰𝗮𝗻 Recently, I started working on an application, with the goal of creating a fast and simple tool that is easy to use yet is effective. And the result... Vulcan. 𝗧𝗿𝘆 𝗩𝘂𝗹𝗰𝗮𝗻: https://lnkd.in/gxPwUMiW 𝗙𝗲𝗮𝘁𝘂𝗿𝗲𝘀: 🔹Very High Inference Speeds: Powered by the Groq LPU™ Inferencing Engine, the response times that I was able to achieve were insane, with speeds as fast as up to 10x compared to traditional GPU based solutions. 🔹Multiple LLMs To Choose From:      ‣ Llama-3.1-8b-instant ‣ Llama-3.1-70b-versatile ‣ Mixtral-8x7b-32768 ‣ Gemma2-9b-it Each one specializes at different tasks. 🔹Document Upload: Upload your documents and ask questions to the LLM based on them. This is made possible through a concept called Retrieval Augmented Generation, otherwise known as RAG. Currently, PFD, TXT and CSV file formats are supported. 🔹Scrape Websites: A feature that analyzes any public websites and lets you ask your questions to the LLM. Currently, analysis of text content is supported. 🔹Code Generation: Just like other ChatBots around, Vulcan is capable of generating code, finding errors in the code, explaining code and much more. This is thanks to the base LLM models that are trained to work on many programming languages and their concepts. So, do try it out and let me know how was your experience. Your feedback is much valued. Thank You! P.S. Vulcan is currently a personal project but, based on your feedback and support, I may decide to create a full-fledged solution that may hopefully outperform all other ChatBots around. Let's see. 𝗠𝘂𝘀𝗶𝗰 𝗖𝗿𝗲𝗱𝗶𝘁𝘀: Jarico (Rail Gimatov) #ChatBot #RAG #Groq #LLAMA #Mistral #Gemma