“I’m building a financial research assistant using @LangChainAI Fully open source. My agent has two powerful search tools: • financial search • web search Financial search: searches specific financial data points across 16K tickers. This lets us precisely search for exact
“❓curiosity This open-source repo uses LangGraph, FastHTML and Tavily to create a Perplexity-like experience Supports different models: gpt-4o-mini by OpenAI llama3-groq-8b-8192-tool-use-preview by Groq llama3.1:latest by Ollama (run your own)
“🕸️ LangGraph Templates 🕸️ It’s never been easier to start creating your own agentic applications. LangGraph Templates are a collection of reference architectures that you can clone, configure and then easily modify. 📝 Read more in the blog post:
“Open Dataset release by @OpenAI! 👀 OpenAI just released a Multilingual Massive Multitask Language Understanding (MMMLU) dataset on @huggingface! 🌍 MMLU test set available in 14 languages, including Arabic, German, Spanish, French,…. 🧠 Covers 57 categories from elementary to
DeepSeek
“DeepSeek 2.5 vs GPT 4o for Coding 21X cheaper than Claude 3.5 sonnet and 17X cheaper than GPT 4o.
“LMSYS Chatbot Arena Rankings Update: DeepSeek-V2.5 has ranked first among Chinese LLMs, outperforming closed-source models like Yi-Large-Preview, Qwen-Plus-0828, and GLM-4-0520. It’s also closely matched with GPT-4-Turbo-2024-04-09 in the arena score. Download the V2.5
Hugging Face
“Exciting update for AI developers! The @huggingface Hub is now more natively integrated into @googlecloud Vertex AI Model Garden. Search through thousands of open Generative AI models from Hugging Face models & deploy them with one click to Vertex AI or GKE. 🤯 What’s new: 🔎
“🛠️ @fdaudens, @huggingface: “Open-Source AI x Journalism: How to Leverage Hugging Face (Intermediate Level)”
Meta/Llama
“Quite an innovative proposal – “Training-Free Long-Context Scaling of Large Language Models” 👨🔧 Dual Chunk Attention (DCA), enabling Llama2 70B to support context windows of more than 100k tokens WITHOUT continual training. 🔥 📌 Dual Chunk Attention (DCA) innovates by
“Introducing LlamaParse Premium 💎 LlamaParse is the best document parser for your LLM applications, and now it’s even better. LlamaParse Premium combines the visual understanding capabilities of SOTA multimodal models with long text/table content extraction capabilities of
“With the release of Llama 3.1 405B, @TogetherCompute built LlamaCoder — an open source web app that can generate an entire app from a prompt. The repo has now been cloned by hundreds of devs on GitHub and starred 2K+ times. More on this project ➡️
“With the release of Llama 3.1 405B, @TogetherCompute built LlamaCoder — an open source web app that can generate an entire app from a prompt. The repo has now been cloned by hundreds of devs on GitHub and starred 2K+ times. More on this project ➡️
“This Paper from Jan-24 can serve a LLaMA-7B with a context length of up to 1 million on a single A100-80GB GPU and up to 10 million on an 8-GPU system 🔥 🗞️ Paper – “KVQuant: Towards 10 Million Context Length LLM Inference with KV Cache Quantization” 📌 The existing problem –
Mistral
“Today we are excited to announce: – Pixtral 12B available on le Chat and la Plateforme – A free tier on la Plateforme – A significant price drop across all our models – An updated Mistral Small Release blogpost:
Qwen
“Alibaba dropped Qwen2.5, a massive new open-source model The Chinese tech giant is continuing to make AI more accessible, with multiple versions for general use, coding, and math The smaller models are also impressively smart (and work in 29+ languages)
“Qwen 2.5 72B Is The Best Open-Source Model In The World The Qwen 2.5 models dropped today, and they have some excellent scores. It beat Llama-405b on Livebench AI (August questions). We will soon release its performance on the September challenge. With an excellent score of
“We have GPT-4 for coding at home! I looked up @OpenAI GPT-4 0613 results for various benchmarks and compared them with @Alibaba_Qwen 2.5 7B coder. 👀 > 15 months after the release of GPT-0613, we have an open LLM under Apache 2.0, which performs just as well. 🤯 > GPT-4 pricing
“Welcome to the party of Qwen2.5 foundation models! This time, we have the biggest release ever in the history of Qwen. In brief, we have: Blog:
“Synthetic data played a crucial role in training Qwen2.5-Coder. Here’s what I figure the process was from the Technical Report: 1. Generation Process: • Used CodeQwen1.5 (predecessor model) to generate large-scale synthetic datasets • Focused on creating diverse, high-quality
“Qwen2.5-72B-Instruct against the opensource models!
Qwen2.5: A Party of Foundation Models! | Qwen
Leave a Reply