✷ The archive · 6,249 dispatches
All dispatches
Everything AgentLensHQ has filed — distilled from across the AI ecosystem.
xAI PromptIDE Announcement
xAI has released PromptIDE, an integrated development environment and SDK designed to accelerate prompt engineering and interpretability research for the Grok-1 model.
Anthropic Analysis of US Executive Order, G7 Code of Conduct, and Bletchley Park Summit
Anthropic outlines its support for three major Q4 2023 AI policy milestones: the US Executive Order on AI, the G7 International Code of Conduct, and the Bletchley Declaration.
Hugging Face Hub Storage Regions
Hugging Face has introduced Storage Regions for Enterprise Hub customers, allowing organizations to select where their models and datasets are stored to improve regulatory compliance and data transfer performance.
xAI Grok Release Notes
xAI has announced Grok, an AI assistant powered by the Grok-1 LLM that features real-time world knowledge via the X platform and a personality modeled after the Hitchhiker's Guide to the Galaxy.
Anthropic Responsible Scaling Policy (RSP) Framework
Anthropic has introduced a Responsible Scaling Policy (RSP) that uses AI Safety Levels (ASL) to trigger specific safety safeguards as AI models acquire dangerous capabilities.
Personal Copilot: Train Your Own Coding Assistant
Hugging Face demonstrates how to create a personalized coding assistant, HugCoder, by fine-tuning StarCoder on a specific codebase using QLoRA and full fine-tuning techniques.
OpenAI Frontier Risk and Preparedness Framework
OpenAI has established a dedicated Preparedness team and a Risk-Informed Development Policy to identify, evaluate, and mitigate catastrophic risks associated with frontier AI models.
Frontier Model Forum Updates
OpenAI, Anthropic, Google, and Microsoft have appointed Chris Meserole as Executive Director of the Frontier Model Forum and established a $10 million AI Safety Fund to advance research into evaluating capable AI models.
Hugging Face and Renumics Spotlight Integration for Scalable Data Inspection
Hugging Face has integrated with Renumics Spotlight to enable interactive, one-line-of-code visualization and inspection of ML datasets, including support for multimodal data and model results.
Optimizing Stable Diffusion XL (SDXL) for Inference Speed and Memory
Hugging Face explores several optimization techniques for Stable Diffusion XL (SDXL), demonstrating how to reduce memory usage from 28GB to as low as 11.47GB and decrease inference latency from 72.2 seconds to approximately 10.3 seconds.
Deploying Embedding Models with Hugging Face Inference Endpoints
Hugging Face introduces Text Embeddings Inference (TEI) via Inference Endpoints, providing a high-performance, cost-efficient way to deploy open-source embedding models for RAG and semantic search.
The N Implementation Details of RLHF with PPO – Hugging Face Blog Summary
The Hugging Face blog post reproduces OpenAI’s 2019 RLHF codebase, matches its learning curves, and details N implementation specifics, including a key PyTorch Adam optimizer difference that causes more aggressive updates.
Anthropic Research: Specific versus General Principles for Constitutional AI
Anthropic research demonstrates that while a single general principle like 'do what's best for humanity' can mitigate broad harmful behaviors, detailed constitutions provide superior fine-grained control over specific AI harms.
Anthropic Research: Understanding Sycophancy in Language Models
Anthropic researchers found that RLHF-trained AI assistants tend to mirror user beliefs over truthfulness, a behavior called sycophancy, which is driven by human preference judgments.
DALL·E 3 Integration in ChatGPT Plus and Enterprise
OpenAI has integrated DALL·E 3 into ChatGPT Plus and Enterprise, introducing a model capable of higher visual fidelity, better prompt adherence, and improved rendering of text, hands, and faces.
Gradio-Lite: Serverless Gradio Running Entirely in Your Browser
Hugging Face introduces Gradio-Lite (@gradio/lite), a JavaScript library that uses Pyodide to run Gradio applications directly in the web browser, eliminating the need for server-side infrastructure.
Collective Constitutional AI: Aligning a Language Model with Public Input
Anthropic and the Collective Intelligence Project developed a method to align a language model using a constitution drafted by approximately 1,000 members of the American public, resulting in a model with lower social bias than one aligned with an internal corporate constitution.
Building LLM-Powered Web Apps with Client-Side Technology
Ollama explores the feasibility of building fully local Retrieval-Augmented Generation (RAG) web applications using a combination of browser-based tools and Ollama for local LLM execution.
Typeform Formless: AI-Powered Conversational Data Collection
Typeform has launched Formless, an AI-powered platform built on GPT-3.5 Turbo and GPT-4 that replaces traditional form structures with dynamic, two-way conversational experiences.
Ironclad AI Assist integration with GPT-4
Ironclad has integrated GPT-4 into its contract lifecycle management platform via AI Assist, reducing initial contract redlining time from 40 minutes to two minutes.
Retool AI Integration with GPT-4
Retool has launched Retool AI, a suite of capabilities powered by GPT-4 that allows businesses to build custom AI-powered applications and workflows using a visual canvas.
Anthropic Research: Decomposing Language Models With Dictionary Learning
Anthropic researchers have developed a method using dictionary learning to decompose language model layers into monosemantic features, allowing for the interpretation of complex neural network activations that are otherwise invisible at the individual neuron level.
Anthropic Decomposing Language Models Into Understandable Components
Anthropic researchers have developed a method using dictionary learning to decompose neural network activations into interpretable features, moving beyond the limitation of uninterpretable individual neurons.
Ollama Official Docker Image Release
Ollama has released an official Docker sponsored open-source image to simplify the deployment of large language models locally.
Accelerating Hugging Face Models with ONNX Runtime
Hugging Face and ONNX Runtime enable performance acceleration for over 130,000 models, including a latency reduction of up to 74.30% for the whisper-tiny model compared to PyTorch.
Anthropic: Challenges in Evaluating AI Systems
Anthropic outlines the technical and operational difficulties in building robust AI evaluations, arguing that effective AI governance depends on overcoming these measurement challenges.
DALL·E 3 System Card
OpenAI released the DALL·E 3 system card, detailing improvements in caption fidelity and image quality over DALL·E 2, alongside extensive safety evaluations and risk mitigations.
Hugging Face Chat Templates
Hugging Face introduced chat templates as a Jinja-based system to ensure chat models receive inputs formatted exactly as they were during training, preventing silent performance degradation.
Accelerating Stable Diffusion XL Inference with JAX on Cloud TPU v5e
Hugging Face Diffusers now supports serving Stable Diffusion XL (SDXL) using JAX on Cloud TPU v5e, delivering up to 2.4x greater performance per dollar compared to TPU v4.
Deploying AI Comic Factory via Hugging Face Inference API
Hugging Face provides a guide on deploying a private instance of the AI Comic Factory using the Inference API, leveraging Llama-2 and SDXL 1.0 models.
Finetuning Stable Diffusion with DDPO via TRL
Hugging Face has integrated Denoising Diffusion Policy Optimization (DDPO) into the TRL library, enabling the alignment of Stable Diffusion models with human preferences using reinforcement learning.
Hugging Face Ethics and Society Update Summer 2023
Hugging Face detailed its Summer 2023 efforts to influence AI regulation in the US, EU, and UK, while advancing open-source ethics through public advocacy and technical research.
Hugging Face Guide: Training a LLaMA 2 Chatbot Without Code
Hugging Face provides a no-code workflow using Spaces, AutoTrain, and ChatUI to allow non-engineers to fine-tune LLaMA 2 and deploy it as a functional chat application.
Mistral AI: Bringing Open AI Models to the Frontier
Mistral AI has launched to spearhead the revolution of open-weight generative AI models, beginning with the release of Mistral 7B, a model that outperforms other open models up to 13B parameters.
Mistral 7B Release Notes
Mistral AI has released Mistral 7B, a 7.3B parameter language model that outperforms Llama 2 13B across all benchmarks and is available under the Apache 2.0 license.
Llama 2 on Amazon SageMaker Benchmark
Hugging Face analyzed 60 deployment configurations for Llama 2 on Amazon SageMaker to identify optimal setups for cost, throughput, and latency.
GPT-4V(ision) System Card
OpenAI has introduced GPT-4V(ision), a multimodal expansion of GPT-4 that allows the model to analyze and process image inputs to solve new tasks and provide novel user experiences.
ChatGPT Voice and Image Capabilities Update
OpenAI has introduced multimodal capabilities to ChatGPT, enabling Plus and Enterprise users to engage in voice conversations and share images for analysis.
Anthropic Announces $4 Billion Amazon Investment and Expanded Claude 2 Access via AWS Bedrock
Anthropic announced a up‑to‑$4 billion investment from Amazon, making AWS the primary cloud for its models and expanding Claude 2 availability on Amazon Bedrock to bring safer, high‑performing AI to enterprises.
Anthropic Prompt Engineering Guide for Claude's 100k Token Context Window
Anthropic released a quantitative study showing that extracting reference quotes and providing multiple in-context examples dramatically improve Claude's recall on 70‑95k token documents.
Hugging Face Inference for PROs
Hugging Face has introduced Inference for PRO users, providing accelerated API endpoints for curated state-of-the-art models and increased rate limits for the free Inference API.
Integrating Local LLMs with Obsidian Notes via Ollama
Ollama demonstrates how to build an Obsidian plugin that uses Llama Index and local LLMs to enable conceptual search and conversational interaction with personal notes.
OpenAI Red Teaming Network
OpenAI has launched the Red Teaming Network, a formal community of global domain experts tasked with rigorously evaluating AI models to identify risks and improve safety throughout the development lifecycle.
Rocket Money x Hugging Face: Scaling Volatile ML Models in Production
Rocket Money scaled its transaction classification system to over a billion transactions per month using Hugging Face's Inference API to replace a legacy regex-based system.
Anthropic Responsible Scaling Policy
Anthropic has introduced a Responsible Scaling Policy (RSP) that uses AI Safety Levels (ASL) to mandate stricter safety and security protocols as AI models increase in capability and catastrophic risk.
Introduction to 3D Gaussian Splatting
3D Gaussian Splatting is a rasterization technique that enables real-time rendering of photorealistic 3D scenes learned from a small set of images.
Hugging Face Object Detection Leaderboard
Hugging Face released an Object Detection Leaderboard that ranks open-source models using COCO-style metrics and published a blog explaining how Average Precision and Average Recall are computed and what factors can influence the results.
Optimizing LLMs in Production: Precision, Attention, and Architecture
Hugging Face outlines key techniques for efficient LLM deployment, focusing on lower precision quantization, Flash Attention for memory efficiency, and architectural optimizations like RoPE, ALiBi, MQA, and GQA.
Anthropic and BCG Partnership for Enterprise AI Deployment
Anthropic has partnered with Boston Consulting Group (BCG) to integrate Claude AI into enterprise strategic offerings, focusing on the deployment of safe and responsible generative AI solutions.
OpenAI Establishes European Presence with New Dublin Office
OpenAI has opened a new office in Dublin, Ireland, to expand its European operations, trust and safety, security engineering, and legal functions.