Notes from projects I build and problems I investigate.
Posts
- A Local LLM Harness for Building and Testing a Browser Game on a Radeon 780M
- Generating bilingual Open Graph images locally with Qwen-Image
- Splitting LTX-2.5 text encoding: 21 minutes to 37 seconds
- Qwen-Image-2512 on the 780M: why ‘Qwen-Image-2.0’ isn’t open-weight
- A local 27B coding agent on the 780M: why it froze, and what unstuck it
