‹ BackHN Continuity

Thread

Run Qwen 3.8 Flash Next (125B) on consumer hardware (RTX 4090) at 100T/s

806 points · 356 comments · snehesht

  1. Neywiny · · focus · HN ↗
    I don't like that some configuration is fine via arguments and others by environment variable. I've noticed LLMs like doing this. And even more, like hallucinating such things. To me the advantage of AI coding is that the boilerplate of command line arguments and passing them around becomes trivial instead of tedious.
    1. halJordan · · focus · HN ↗
      100% not an llm problem. Llama.cpp, which only recently started taking large amounts of ai code has had this problem for years.
Open on Hacker News to reply ↗

Unofficial Hacker News client; not affiliated with Y Combinator.