Mervin Praison
@mervinpraison
How to Create Serverless RAG Application? LLM : Llama 3.1 using Groq Inc Serverless Vector Database: Neon - Serverless Postgres Backend: Flask Python Framework: phidata youtube.com/watch?v=VNILuEโฆ AI at Meta Ashpreet Bedi
OpenAI's Strawberry o1: World's best AI released! Test Results โ๏ธ Reasoning-focused AI ๐ OpenAI's o1 Outperforms GPT-4o ๐ PhD-level task accuracy ๐ง Excels at complex problems ๐งโ๐ป Specialised for coding ๐ก Self-fact-checking Sub: youtube.com/watch?v=43PtzKโฆ OpenAI OpenAI Developers
How does OpenAI train the Strawberry๐ (o1) model to spend more time thinking? I read the report. The report is mostly about ๐ธ๐ฉ๐ข๐ต impressive benchmark results they got. But in term of the ๐ฉ๐ฐ๐ธ, the report only offers one sentence: "Through reinforcement learning, o1
AI News 0:00 - ๐ Introduction 0:37 - ๐ฃ๏ธ Empathetic Voice Interface (EVI 2) Hume 1:46 - ๐ฆ Llama Omni 8B AI at Meta 2:24 - ๐ Jina AI reader - HTML to markdown LM 2:33 - ๐ Fish Speech: open-source TTS Fish Audio 3:01 - ๐ท OCR2 Philipp Schmid 3:23 - ๐๏ธ Mistral AI Pixtral 12B
AI News from Mervin Praison hubs.la/Q02PV5YF0
OpenAI o1 in ChatLLM & more features Abacus.AI bit.ly/abacusai-chatlโฆ ๐ฅ๏ธ Multi-model AI interface ๐จ AI image generation ๐ป Code creation and testing ๐ AI-powered presentations ๐ URL scraping and analysis ๐ธ Screenshot interpretation ๐ Custom knowledge integration ๐ฎ UI
Google NotebookLM: ANYONE Can Create Podcast in 5 mins using AI ๐ Upload and summarise documents โ AI-powered Q&A on uploaded content ๐๏ธ AI-generated podcasts with human voices ๐ Google Drive integration ๐ Virtual research assistant capabilities ๐๏ธ Organize and synthesize
unexployed WIP alแปa Mervin Praison is a machine for simple, beginner-friendly YouTube videos that are always bleeding edge He also supplies a ton of his matching code (for free!) that you can clone and play around with
Llama 3.2 Released by AI at Meta ๐ Model Sizes: 1B, 3B, 11B, 90B ๐ฑ On-device: 1B, 3B | Multimodal: 11B, 90B ๐ 1B and 3B models support 128K Context Length ๐ 90B model outperforms GPT-4 in some benchmarks. ๐ Available on cloud providers and model APIs. ๐ 15% fewer tokens