“Had a series B company come to us to build an agent. They had built a version of it in LangGraph over the past 3-weeks. In 5-minutes, @braelyn_ai had a prototype built with almost all the functionality they needed using AgentStack from @AgentOpsAI. Great work @braelyn_ai
“Since our first version of Hermes was released over a year ago, many people have asked for a place to experience it. Today we’re happy to announce Nous Chat, a new user interface to experience Hermes 3 70B and beyond.
“Introducing Oasis: the first playable AI-generated game. We partnered with @DecartAI to build a real-time, interactive world model that runs >10x faster on Sohu. We’re open-sourcing the model architecture, weights, and research. Here’s how it works (and a demo you can play!):
SmolLM2
The Gap Between Open and Closed AI Models Might Be Shrinking | TIME
How to build an Open Source NotebookLM: PDF to Podcast – https://docs.together.ai/docs/open-notebooklm-pdf-to-podcast
“Just tested @Argilla’s new data annotation feature – it’s a game changer for AI project quality. Upload CSVs, work with published datasets, or improve existing ones directly on HuggingFace Hub. Setup took < 2 minutes, no code needed (see example below where I selected a dataset
“”Modern open-source time-tracking app”
“”Infisical is the open-source secret management platform: Sync secrets across your team/infrastructure, prevent secret leaks, and manage internal PKI”
Near plans to build world’s largest 1.4T parameter open-source AI model
“”Infisical is the open-source secret management platform: Sync secrets across your team/infrastructure, prevent secret leaks, and manage internal PKI”
Deep Imagination Research | NVIDIA
Cosmos Tokenizer: A suite of image and video neural tokenizers (upscalers)
“✨SCIPE✨ – Systematic Chain Improvement and Problem Evaluation TLDR: It helps you find underperforming nodes in LLM chains New research out of Berkeley (@sh_reya and Ankush Garg) that integrates seamlessly with LangGraph “In this post, we introduce SCIPE, a lightweight, yet
“Build an Agent-Native application with LangGraph & CoAgents in a few minutes🦜🪁🚀 🎥New Tutorial by @ataiiam
LangChain on X: “New to LangSmith? Check out our 7-part video series that walks through the basics of LangSmith for LLM testing & observability. Learn how to: 🔍 Set up tracing for visibility into your LLM app 🛝 Build & refine prompts in the Playground 🛠️ Curate datasets & run robust https://t.co/SGsMkb7fJx” / X – https://x.com/LangChainAI/status/1854563857525805125
“Build better RAG systems by using the LLM to intelligently expand and refine your retrieved context 🧠💡 In this guest blog post, learn how to build a Context Refinement Agent that improves RAG responses for complex queries by using an agent to examine the retrieved chunks for
“AI agent orchestration framework for building LLM apps with RAG and multiple AI Agents in Python. 100% opensource.
“Proposes a self-evolving online curriculum RL framework to bridge the gap between open and proprietary LLM-based web agents. It improves the success rate of Llama-3.1-8B from 4.8% to 42.4%, and from 6.1% to 43% for GLM4-9B. The open models significantly surpass the performance
“🌟 Excited to open-source SELA, a powerful experimentation system integrating MCTS with LLM agents. Across 20 datasets, SELA achieves a 75% win rate against AIDE (OpenAI’s top pick in MLE-Bench) and beats traditional AutoML methods developed over years. 💻 Code:
“Open source is the path to autonomous AI-powered Software Engineers. Several companies are trying to build “AI Software Engineers”, but now that we have an open-source framework, things will get crazy! → (Everyone who’s mentioned they’d want a coding agent but can’t use what’s
Meta/Llama
Open Source AI Can Help America Lead in AI and Strengthen Global Security | Meta – https://about.fb.com/news/2024/11/open-source-ai-america-global-security/
Meta’s former hardware lead for Orion is joining OpenAI | TechCrunch
Meta is using more than 100,000 Nvidia H100 AI GPUs to train Llama-4 — Mark Zuckerberg says that Llama 4 is being trained on a cluster “bigger than anything that I’ve seen” | Tom’s Hardware
“Join us in San Francisco (or online) this weekend for a Llama Impact Hackathon! Teams will be spending two days building new ideas and solutions with Llama 3.1 + Llama 3.2 vision and on-device models. Three challenge tracks for this hackathon: 1️⃣ Expanding Low-Resource Languages” / X
“Build a chat UI for your LLM app in minutes with LlamaIndex chat-ui! 🚀💬 This React component library offers: ➡️ Pre-built chat components (message bubbles, input fields) ➡️ Tailwind CSS customization ➡️ Easy integration with LLM backends like @vercel AI Key features: •” / X
“Ollama 0.4 is released with support for Meta’s Llama 3.2 Vision (11B and 90B) models. @AIatMeta See what it can do: 👇👇👇
LlamaPReview · GitHub Marketplace
“@AIatMeta Examples of running Llama 3.2 vision: Reading handwriting
Mistral
Mistral launches a moderation API | TechCrunch
“We @MistralAI just released the moderation API and the batch API. Details in 🧵:
Mistral Moderation API | Mistral AI | Frontier AI in your hands
“@MistralAI We introduced a new Mistral Moderation model, a classifier model based on Ministral 8B. It enables our users to detect harmful text content along 9 policy dimensions. Blog: https://x.com/sophiamyang/status/1854622256993059220
Leave a Reply