Show HN: Woolly, a local mem0-compatible memory API for AI agents
Woolly is a locally run memory API compatible with mem0, giving AI agents a self-hosted long-term memory.
Why this: Covers Agent and API · Published within 24 hours
20 items · All topics →
Woolly is a locally run memory API compatible with mem0, giving AI agents a self-hosted long-term memory.
Why this: Covers Agent and API · Published within 24 hours
LangChain raised $300M in new funding to help companies own their agent capabilities through LangSmith.
Why this: Official source · Covers Agent and API · Published within 24 hours
ROCKET is a training-free model compression method that uses calibration-guided truncation to achieve leading performance at high compression ratios.
Why this: Covers model and API · Relevant to this topic
After an API migration, code fixed by an LLM can run without errors yet produce wrong results — and standard release checks miss these silent failures entirely, the research finds.
Why this: Covers LLM and release · Relevant to this topic
This distributed asynchronous RL framework keeps vision-language-action model training apace with the rapid evolution of embodied AI.
Why this: Covers model and multimodal · Relevant to this topic
This work enables cloud and edge LLMs to learn collaboratively by accelerating personalized bilevel optimization, supporting distributed training for physical-world intelligence.
Why this: Covers model and LLM · Relevant to this topic
This explainable judge scores text-to-audio generation and explains its ratings, fixing metrics that give numbers without reasons.
Why this: Covers API · Relevant to this topic
This study tests whether projection-derived masks can replace full 3D segmentation for cervical spine fracture recognition, aiming to make automatic CT interpretation faster and cheaper.
Why this: Covers API · Relevant to this topic
CPGRec+ balances personalization and diversity for the fast-changing game industry, making game recommendations better matched to players' tastes.
Why this: Covers API · Relevant to this topic
With Gemini Decision 1, Google enters the decision-model space alongside Jev and OpenAI; the API mirrors Jev's design and covers three question types.
Why this: Covers model and Gemini · Relevant to this topic
This official AWS emulator, written in Rust, lets you test services locally without provisioning a paid API key.
Why this: Covers API · Relevant to this topic
Singapore Airlines put OpenAI models and APIs to work on customer support, travel discovery, and staff efficiency — and its Kris chatbot's satisfaction scores doubled.
Why this: Official source · Covers model and API
A 'Do not ask again' prompt ended up making passwords, API keys, and other secrets public, so Claude Code users should review their settings.
Why this: Covers Claude and prompt · Relevant to this topic
OpenAI's Control Room keeps watch over model workloads spread across its research infrastructure, and can halt them fast if a safety issue comes up.
Why this: Official source · Covers model and API
With the OpenAI API, Kuken searches 70 years of cooling tower drawings — what once took an engineer nearly an hour now takes 4.5 minutes to find the drawing for a repair.
Why this: Official source · Covers API
This release lets agents work directly on sites without APIs and sign in themselves, while background subagents handle long-running tasks.
Why this: Official source · Covers Agent and API
The new version generates images faster, adds GIF creation and sharper editing, and brings fresh controls to ChatGPT, Codex, and the API.
Why this: Official source · Covers GPT and API
In this release, ollama launch claude honors the output token limit of the model, keeping cloud models clear of the 128k default cap, while cloud web search responses also improve.
Why this: Covers model and Claude · Relevant to this topic
OpenAI describes how Habitat was scaled with dependable multi-tenancy, worldwide caching, and Azure Cosmos DB to deliver 12x faster reads.
Why this: Official source · Covers GPT and API
The Decisions API brings quick, focused, low-cost decision-making to any application.
Why this: Official source · Covers API