How it works
Google Open-Sources Harness² , a Framework That Improves AI Agents Without Retraining Models
A new open-source tool from Google Research lets agents get better at tasks by tweaking their setup instead of their core model.
Tag
#open-source
12 stories.
How it works
New AI plugin lets Claude design PCB boards inside KiCad 10
Konnect is a free open-source plugin that lets an AI assistant place components, route circuits and export manufacturing files directly inside KiCad.
New modelsFLUX.3 Image Klein Arrives in Hugging Face DiffusersHerald of AI
New models
FLUX.3 Image Klein Arrives in Hugging Face Diffusers
The 6-billion-parameter image-to-image model from Black Forest Labs is now available through the popular open-source library.
How it worksClawLoop AI agents that teach themselves to get betterHerald of AIFigure: Yang et al., arXiv:2610.14118, CC BY 4.0
How it works
ClawLoop AI agents that teach themselves to get better
A new training framework lets AI agents improve on their own without a stronger teacher model guiding every step.
How it worksLeCun's Startup Unveils H-JEPA A Hierarchical World Model for Visual PlanningHerald of AI
How it works
LeCun's Startup Unveils H-JEPA A Hierarchical World Model for Visual Planning
AMI releases its first research with full code and weights open-sourced, showing that stacking prediction layers dramatically improves long-term planning in visual tasks.
New modelsAnthropic launches free AI scanner to protect open-source codeHerald of AI
New models
Anthropic launches free AI scanner to protect open-source code
The new tool automatically finds security flaws in open-source projects so maintainers can patch them before attackers exploit them.
New modelsCompressed 27B Model Beats Full Version at Tool Calling in Under 8 GBHerald of AI
New models
Compressed 27B Model Beats Full Version at Tool Calling in Under 8 GB
A heavily compressed version of Qwen3.8-27B runs on everyday hardware and outperforms the original on agent skills.
How it worksGoogle DeepMind Launches EmbeddingGemma 2 for On-Device Multimodal SearchHerald of AI
How it works
Google DeepMind Launches EmbeddingGemma 2 for On-Device Multimodal Search
The new 740M-parameter model can search text, code, images, video and audio on your phone or laptop without sending data to the cloud.
How it worksArchitect Launches Liquid Inference, an Auction Marketplace for LLM ComputeHerald of AI
How it works
Architect Launches Liquid Inference, an Auction Marketplace for LLM Compute
A financial-tech company is applying exchange-style auction mechanics to LLM inference, letting providers bid for every request.
New modelsPerplexity Releases Two Multimodal Embedding Models for Edge and CloudHerald of AI
New models
Perplexity Releases Two Multimodal Embedding Models for Edge and Cloud
The 9B model scores 92.4% on agentic PDF QA, while the 0.6B variant runs on laptops for fast, cheap queries.
How it worksMeta Open-Sources Rebalancer A Solver That Handles 40M Assignment Problems DailyHerald of AI
How it works
Meta Open-Sources Rebalancer A Solver That Handles 40M Assignment Problems Daily
A C++ library that decides where objects belong — servers, tasks, traffic — is now free for everyone.
How it worksLiquid AI Opens Up d1 Decision Models for Real-Time, Edge AI ChoicesHerald of AI
How it works
Liquid AI Opens Up d1 Decision Models for Real-Time, Edge AI Choices
Two new open-weight multimodal decision models return answers with zero output tokens — built for speed and routing tasks on everything from servers to edge chips.