• Home
  • Technology
  • Gaming
  • Entertainment
  • World & Business
  • Science
  • Sports
  • AI
HomeTechnologyGamingEntertainmentWorld & BusinessScienceSportsAI
    • Home
    • Technology
    • Gaming
    • Entertainment
    • World & Business
    • Science
    • Sports
    • AI
    AI

    Running FLUX.2 on AMD MI300X with the Luminal compiler

    Luminal says cheaper floating-point operations help it lower the cost per image, an approach one of its engineers has written up.

    Y CombinatorYC
    Paul BuchheitPB
    Dalton CaldwellDC
    4 Sources, ,

    TLDR

    Luminal says it runs large-scale FLUX.2 diffusion workloads on AMD MI300X through its compiler. The company credits cheaper floating-point operations with driving down its cost per image and points to an engineer’s write-up explaining the approach.

    Combined views

    10.5K

    4 Sources, first seen 20d ago

    Combined views

    10.5K

    4 Sources, first seen 20d ago

    76 likes
    20d ago
    first seen 20d ago
    76 likes
    13 comments
    7 saves
    48 reposts
    13 comments
    7 saves
    48 reposts

    Sentiment

    Positive——Negative

    Summary

    Not enough discussion yet.

    No sentiment analysis available yet.

    Today's Rank

    —

    Not ranked yet

    Today's Rank

    —

    Not ranked yet

    Sentiment

    Positive——Negative

    Summary

    Not enough discussion yet.

    No sentiment analysis available yet.

    4 Sources

    Luminal@luminal_aiWe run large-scale FLUX.2 diffusion workloads on AMD MI300X through the Luminal compiler. Our engineer Reis wrote up how we drive down the cost per image by utilizing cheaper flops.20d
    Paul Buchheit@paultooRT @luminal_ai: We run large-scale FLUX.2 diffusion workloads on AMD MI300X through the Luminal compiler. Our engineer Reis wrote up how w…20d
    Dalton Caldwell@daltoncRT @luminal_ai: We run large-scale FLUX.2 diffusion workloads on AMD MI300X through the Luminal compiler. Our engineer Reis wrote up how w…20d
    Y Combinator@ycombinatorRT @luminal_ai: We run large-scale FLUX.2 diffusion workloads on AMD MI300X through the Luminal compiler. Our engineer Reis wrote up how w…19d

    4 Sources

    Luminal@luminal_aiWe run large-scale FLUX.2 diffusion workloads on AMD MI300X through the Luminal compiler. Our engineer Reis wrote up how we drive down the cost per image by utilizing cheaper flops.20d
    Paul Buchheit@paultooRT @luminal_ai: We run large-scale FLUX.2 diffusion workloads on AMD MI300X through the Luminal compiler. Our engineer Reis wrote up how w…20d
    Dalton Caldwell@daltoncRT @luminal_ai: We run large-scale FLUX.2 diffusion workloads on AMD MI300X through the Luminal compiler. Our engineer Reis wrote up how w…20d
    Y Combinator@ycombinatorRT @luminal_ai: We run large-scale FLUX.2 diffusion workloads on AMD MI300X through the Luminal compiler. Our engineer Reis wrote up how w…19d