Tech News
Show HN: HN.watch – Videos of all Hacker News posts
Scrimba launched HN.watch, a demo that turns Hacker News posts into AI-generated explainer videos using an HTML-based format, with generation costs around $0.04 per video.
Add Runtime Controls to AI Agents with NVIDIA OpenShell
AI agents can be given a goal, write code, use tools, and keep working as new information becomes available. This opens the door to applications.
Grok 4.7 is now available on Amazon Bedrock
xAI's Grok 4.7 is now available on Amazon Bedrock: a frontier model for coding, long-running agents, and knowledge work.
Quoting @joedaroo
To say that we were surprised at the jump and suddenness of the capabilities of our models when it came to “cyber” or “swarming” or “message boards” or anything else related to the incidents is an understatement.
openai/openai-python v3.20.0
3.20.0 Features api: add Agents credential and session options.
GitHub Repos
D4Vinci/Scrapling
🕷️ An adaptive Web Scraping framework that handles everything from a single request to a full-scale crawl!
# Python# ai# ai scrapingopendatalab/MinerU
Transforms complex documents like PDFs and Office docs into LLM-ready markdown/JSON for your Agentic workflows.
# Python# ai4science# document analysistrpc-group/trpc-agent-go
A Go framework for building production agent systems with graph workflows, tools, memory, A2A, AG-UI, MCP, evaluation, and observability.
# Go# a2a# a2a protocolResearch Papers
Denoising Multi-Robot Trajectories
This paper builds on a diffusion-denoising planner called D4orm to create several multi-robot trajectory planners that work in different settings, like online control and distributed teams. They show these planners find good paths faster than other sampling methods and test...
ABC-Align: Prediction-Powered Alignment with Adaptive Bias Control
This paper proposes ABC-Align, a method to align language models using many cheap AI-generated labels plus a few human labels, automatically adjusting for bias. It aims to reduce the need for expensive human preference data while keeping alignment quality high.
Reasoning with Continuous Latent Diffusion
Continuous diffusion generates complete reasoning solutions through iterative refinement in latent space. We introduce Latent Flow Reasoning Models (LFRMs), an ELF-based training and inference recipe.