Thread

Shapelearn Qwen 3.8 27B (13.1 GB VRAM)

104 points · 39 comments · syntaxing

npodbielski · 1789738210 · focus · HN ↗
Well I tested it on 7900XTX with the same prompts and their draft model gave me about 30t/s. Their own snippet of code with regular MTP model gave me 60t/s.
Also model with their draft answered incorrectly. With MTP it answered correctly.
Question was: "Does MikroTik CRS312-4C+8XG-RM have combo ports?". The answer is Yes.
1. npodbielski · 1789742345 · focus · HN ↗
  When I changed the number of draft tokens to 3 in both, it helped and they Draft is actually performing a bit better:
  - draft: 67.17
  - MTP: 64.18
  Why they used those examples? Seems strange.

Unofficial Hacker News client; not affiliated with Y Combinator.