Skip to content

Recommendations & Tips

Run Qwen 3.6 35B-A3B locally in OpenCode

SoftwareQwen3.6-35B-A3B (local) in OpenCode

model okay use on open code run it locally and run it on open code it is really real Awesome. And it's so fast.

Episode 258: Perceptron (episode page, opens in a new tab) April 24, 2026, 2:40 p.m. CT 01:38:09 (play audio from this timestamp, opens in a new tab)

Voice-matched

Summary

Dave urges Adam to try the Qwen 3.6 35B A3B mixture-of-experts model running locally in the OpenCode agent. It's very good and so fast that output arrives almost before you hit Enter.

Not in the audio. 'Quinn' in the transcript is Qwen (Alibaba). The model name is spoken in a lone Dave cue and an adjacent Unknown cue just before the quote. Context confirms Dave: he's urging Adam to 'try' it, and Dave uses this model on his DGX Spark in other episodes.