Overview of recent notable open-weight AI model releases, including Nanbeige, Laguna, Motif, Solar Open, and Antares.
Overview of recent notable open-weight AI models including Nanbeige, Laguna, Motif, Solar, and Antares with architectural details.
Martin Fowler shares notes from a Thoughtworks retreat on AI harness engineering and self-hosted models.
Historical comparison between 18th-century suppression of Diderot's Encyclopédie and modern debates over open AI models, exploring knowledge access politics.
Tutorial on setting up a local coding agent using open-weight LLMs and open-source tools as an alternative to Claude Code and Codex.
Tutorial on setting up a local coding agent using open-weight LLMs and open-source tools as an alternative to Claude Code and Codex.
Testing local open-weight LLMs like Qwen-Code and Codex in coding harnesses, comparing token use and task success.
Testing local open-weight LLMs like Qwen-Code and Codex in coding harnesses, comparing token efficiency and performance.
Analysis of AI coding agent costs in 2026, comparing model performance, harness efficiency, and pricing between US and Chinese models.
A talk on implementing LLM architectures from scratch in Python and PyTorch, covering workflow for analyzing new models.
A talk on implementing LLM architectures from scratch in Python and PyTorch, focusing on tracing design changes in open-weight models.
A workflow for understanding open-weight LLM architectures using config files and code from Hugging Face.
A workflow for understanding new open-weight LLM architectures using config files and code from Hugging Face.
A gallery showcasing and comparing architecture diagrams and technical details of recent open-weight Large Language Models (LLMs).
A gallery showcasing architecture diagrams and technical details for recent open-weight Large Language Models (LLMs).
Analysis of China's Kimi K2 Thinking AI model, a low-cost, open-weight model challenging US dominance in reasoning and agentic tasks.
A technical analysis of DeepSeek V3.2's architecture, sparse attention, and reinforcement learning updates, comparing it to other flagship AI models.
