Built a landing page with Qwen 3.8 in 20 minutes.
100% local. Not frontier-model fast, but damn it's good and more than fast enough for something that costs $0. Stack: Qwen3.8-27B (MTPLX-optimized) served by MTPLX on an M4 Max, ~52 tok/s, with pi as the coding agent.
Qwen 3.8 27B on Atomic Agent was 2x faster than on Hermes and Prime agents!
Qwen 3.8 Max Full COURSE 1 HOUR (Build & Automate Anything)
Full local AI music animation with a bit of after effects.
Reference Images - Anima animation - minimax H3 music - minimax music 3 llm prompts - Qwen 3.8 27B All done with a single home compute.🥳
You have reached the end of the archive
All of qwen38