Running qwen 3.8 flash next on one amd halo box.
Looks decent!
Quick demo showing the vision capabilities of Qwen 3.8-27B 👇
I gave it a crinkled, scanned-looking document containing PII. It extracted the text, analyzed the document, and identified the PII ✅ This little model is a beast 😄 And it was ticking along at 272 tok/sec on
Little bit of physical harness work with Qwen 3.8 27b
Testing Qwen 3.8 Flash IQ3_S (GSQ RCO) on a single RTX 3090, this time paired with DDR5-5200 MT/s memory.
You have reached the end of the archive
All of qwen38