• 0 Posts
  • 66 Comments
Joined 3 years ago
cake
Cake day: June 18th, 2023

help-circle
  • I see i can still order 96gb, for about 100k nok. That money would give me 2x dxg sparks - although with lower memory bandwidth - but 128gb each. And you don’t need ram for gui etc.

    So 4 sparks can load deepseek r1 or v3 in nvfp4 and have have about 4kTops

    For MoE this is acceptable, for dense it ism’t. Not sure how it fares on diffusion models.

    They was kings when they had 256 and 512 versions but they stopped that this spring












  • Rough estimate from my personal connections only. Some work places where ai is not possible, but all that have made an effort report good code. You need to work with what it is - a word generator that sometimes gives correct results. Make it research and not trust training. Never let it do things on its own, require a plan and reason. Make it evaluate its own work/plan.

    Most issues i have stem from models beeing too eager. Restrain them and remove the “i can do this next…”behaviour.

    Context is king - so proper mcp and documentation that is agent facing. I use serena as i can get lsp for yaml, markup and keep these docs like that