TencentCloud /CubeSandbox
Instant, Concurrent, Secure & Lightweight Sandbox for AI Agents.
→ View original sourceaws /agent-toolkit-for-aws
Official, AWS-supported MCP servers, skills, and plugins to help AI agents build on AWS
→ View original source
How to Rank Local LLMs by Cost per Correct Answer (Measured GPU Energy, 8 Ollama Models)
TL;DR: I priced 8 local Ollama models by € per 1,000 correct answers — metered GPU energy ÷ correct answers, on one RTX 3090. gemma4:26b won at 96.9% accuracy for €0.013/1k-correct. The most expensive
→ View original source
I fine-tune small 7B models into single-voice "character modules" instead of prompt-wrapping a persona. ~20 historical/literary voices (Herodotus, Clausewitz, Kafka…), open weights + a free console.
> "Chance, like friction and fog, prevents a commander's plans from flowing along their intended lines. Genius consists chiefly in the skill to turn chance into advantage." That's
→ View original source[Release] HyperspaceDB v3.1.0: We built a Rust-native Spatial AI Engine that uses 50x less RAM than Milvus/Chroma via Matryoshka Cascades and Lorentz Geometry.
Hey everyone! 👋 If you’re building RAG or autonomous AI agents, you’ve probably hit the "Vector DB Wall": flat Euclidean vectors suck at modeling complex hierarchical reasoning, and loa
→ View original source
ShotcreteDepth: A Bi-modal Dataset for Robust Robotic Depth Perception in Shotcrete Construction Environments
We introduce ShotcreteDepth, a bi-modal dataset from the construction domain that captures both an active shotcreting process and general construction environments. The dataset comprises stereo RGB im
→ View original source
Claude Fable 5 Suspended: The AI Shutdown That Exposed the New Battle Over Frontier Models
Claude Fable 5 Suspension Highlights Escalating Tensions in Frontier AI Model Development Anthropic's recent suspension of its Mythos-class model development, alongside geopolitical developments in AI infrastructure and …
→ View original sourcelangbot-app /LangBot
LangBot: A Production-Grade Framework for Multi-Platform Agentic IM Bots LangBot is an enterprise-ready platform designed for the development and deployment of agentic intelligent bots, featuring sophisticated orchestrat…
→ View original sourceanthropics /claude-plugins-official
Anthropic Releases Official Directory for Claude Code Plugins Anthropic has launched an official, managed repository designed to provide a curated collection of high-quality plugins for Claude Code, streamlining the exte…
→ View original sourceI trained a tiny (6M-param) attention-free model you can chat with, generates a sentence in ~5 ms on CPU, no GPU, no pretrained embeddings. Honest writeup.
Efficient Sequence Generation: A 6M-Parameter Attention-Free Model for Premise Synthesis A researcher has developed a lightweight, 5.98M-parameter sequence model capable of generating sentences in approximately 5ms on a …
→ View original sourceWhat I've noticed about running Gemma 4 12B Unified
I'm new to local LLMs. When I learned that Google's Gemma 4 12B fits on my 4060 TI 16 GB, I set it up with Ollama and started playing. (It's worth pointing out that I don't do any coding tasks). I
→ View original source