GV

GeezerVison

The World Wide Web, restored to 800 pixels of glory.

Home|About|Gaming cabinet|Archives|How it works|Starter sites

On this site

  • Sign in
  • ActionsCI/CD and automation
  • CodespacesCloud development environments
  • All Features
  • Documentation
  • Training
  • Certifications
  • Build with Copilot
  • Customize Copilot
  • Integrations
  • Trending repos
  • GitHub Sponsors
  • Developer articles
  • Enterprises

This view

  • Original page
  • New address

GeezerVison is showing github.com โ€” ROCm/TheRock ยท Discussions

ROCm TheRock Discussions

Pinned Discussions

  • scottt

    gfx1151: I have pytorch, pytorch-vision, and hipBLASLt working well enough to run GPT2

    ๐Ÿ’ฌ General ยท scottt
Clear
    Sort by: Latest activity
    Latest activity
    Date created
    Top: Past day
    Top: Past week
    Top: Past month
    Top: Past year
    Top: All
    • Open
    • Closed
    • Locked
    • Unlocked
    • Answered
    • Unanswered
    • Verified
    • All

    Categories, most helpful, and community links

    Categories

    Loading

    Uh oh!

    There was an error while loading. Please reload this page.

    Community links

    • Community guidelines

    Discussions

    • You must be logged in to vote
      ๐Ÿ™

      How to uninstall?

      joshfoss asked Nov 13, 2026 in Q&A ยท Unanswered
      @joshfoss @Sagargupta16
      @joshfoss @Sagargupta16
      2
    • You must be logged in to vote
      ๐Ÿ™

      (ROCm) AMD-GPU-Boost: Unlock Full Performance on Consumer AMD GPUs

      Under Investigation
      Painter3000 asked Aug 24, 2025 in Q&A ยท Unanswered
      @Painter3000 @robertojguerra
      @alain-bkr
      @Painter3000 @robertojguerra
      @alain-bkr
      10
    • You must be logged in to vote
      ๐Ÿ’ฌ

      Experience with APU Ryzen 7 6850H 680m

      robertojguerra started Aug 7, 2026 in General
      @robertojguerra @LuXuxue
      @robertojguerra @LuXuxue
      43
    • You must be logged in to vote
      ๐Ÿ™

      HIP out of memory when there appears to be plenty of memory available

      Under Investigation
      mycomedico asked Aug 25, 2023 in Q&A ยท Unanswered
      @mycomedico @trvon
      @fluidnumericsJoe
      @mycomedico @trvon
      @fluidnumericsJoe
      6
    • You must be logged in to vote
      ๐Ÿ™Œ

      RDNA1 (gfx1010) optimized ComfyUI fork

      Zakhrov started Oct 14, 2026 in Show and tell
      @Zakhrov @robertojguerra
      @Zakhrov @robertojguerra
      1
    • You must be logged in to vote
      ๐Ÿ™Œ

      I did it: one-click local AI music, no NVIDIA needed

      HiImHal started Oct 25, 2026 in Show and tell
      @HiImHal
      @HiImHal
      0
    • You must be logged in to vote
      ๐Ÿ™

      Naive matrix multiplication on gfx1153 and Windows 11

      medvidec asked Oct 21, 2026 in Q&A ยท Unanswered
      @medvidec
      @medvidec
      0
    • You must be logged in to vote
      ๐Ÿ™

      gfx1100 (RX 7900 XTX): rocBLAS f16 GEMM intermittently returns whole NaN output columns. Reproducer inside; looking for other cards to compare

      roverdrubber asked Sep 18, 2026 in Q&A ยท Unanswered
      @roverdrubber
      @roverdrubber
      2
    • You must be logged in to vote
      ๐Ÿ™

      Help - install ROCm in Docker container - Linux Mint Virginia

      Jopp-gh asked Sep 6, 2026 in Q&A ยท Answered
      @Jopp-gh @robertojguerra
      @sbryngelson
      @Jopp-gh @robertojguerra
      @sbryngelson
      15
    • You must be logged in to vote
      ๐Ÿ™

      Support of Radeon on Debian

      bronger asked Aug 4, 2026 in Q&A ยท Unanswered
      @bronger @sbryngelson
      @bronger @sbryngelson
      1
    • You must be logged in to vote
      ๐Ÿ’ฌ

      seeking "linux appliance" working examples w Ryzen + 780M (gfx1103) + onnx

      CJP-work-acct started Sep 4, 2026 in General
      @CJP-work-acct @robertojguerra
      @CJP-work-acct @robertojguerra
      5
    • You must be logged in to vote
      ๐Ÿ™

      Does latest ROCm 6.0 support Radeon 680M๏ผˆgfx1035๏ผ‰ ?

      PennyFranklin asked Feb 27, 2024 in Q&A ยท Unanswered
      @PennyFranklin @robertojguerra
      @BikeHelmet
      @PennyFranklin @robertojguerra
      @BikeHelmet
      10
    • You must be logged in to vote
      ๐Ÿ’ก

      Feature Request: ROCm Support for AMD Ryzen 9 7940HS with Radeon 780M Graphics

      pearsonc started Jun 26, 2024 in Ideas
      @pearsonc @dreirund
      @MGAndreasen
      @pearsonc @dreirund
      @MGAndreasen
      7
    • You must be logged in to vote
      ๐Ÿ™Œ

      Self-contained Navi 3(gfx110x) and Strix Halo (gfx1151) Pytorch Wheels

      scottt started May 19, 2025 in Show and tell
      @scottt @Tyruiop
      @waltercool
      @scottt @Tyruiop
      @waltercool
      167
    • You must be logged in to vote
      ๐Ÿ™Œ

      Success: LTX-2 Video+Audio Generation on Strix Halo (gfx1151) with TheROCk 7.11

      bkpaine1 started Jan 10, 2026 in Show and tell
      @bkpaine1 @powderluv
      @stellaraccident
      @bkpaine1 @powderluv
      @stellaraccident
      9
    • You must be logged in to vote
      ๐Ÿ’ก

      Post-hackathon contribution offer: RDNA4 (gfx1201) LLM inference findings โ€” kernels, ISA probes, profiler gaps, host tuning

      The-Monk started Aug 17, 2026 in Ideas
      @The-Monk @doplxyz
      @The-Monk @doplxyz
      4
    • You must be logged in to vote
      ๐Ÿ’ก

      Design: ccache accuracy for release builds (fingerprint coverage vs remote cache)

      subodh-dubey-amd started Aug 11, 2026 in Ideas
      @subodh-dubey-amd
      @subodh-dubey-amd
      2
    • You must be logged in to vote
      ๐Ÿ’ฌ

      Testing ComfyUI with RDNA1 (Radeon RX 5600M) and with the Vega iGPU

      Zakhrov started Jan 30, 2026 in General
      @Zakhrov @robertojguerra
      @Zakhrov @robertojguerra
      5
    • You must be logged in to vote
      ๐Ÿ™Œ

      ROCm Issue Agent โ€” an experimental local diagnostic and MCP toolkit

      davidchen-rocm started Aug 4, 2026 in Show and tell
      @davidchen-rocm
      @davidchen-rocm
      0
    • You must be logged in to vote
      ๐Ÿ™Œ

      [Showcase] Native HIP/C++ Z-Image-Turbo optimized for RX 7900 XTX / gfx1100

      uulong950 started Aug 4, 2026 in Show and tell
      @uulong950
      @uulong950
      0
    • You must be logged in to vote
      ๐Ÿ™

      bge-m3 on ROCm 7.2 / gfx1151 (RDNA 3.5): pooling precision issue + DMA SEGV on 7.0 + missing FP8/BF16 support

      slamdunk111 asked Jul 26, 2026 in Q&A ยท Unanswered
      @slamdunk111 @RenSolvyn
      @slamdunk111 @RenSolvyn
      1
    • You must be logged in to vote
      ๐Ÿ™Œ

      Release: Community port of tiny-cuda-nn and gsplat for RDNA4 (gfx1201) / ROCm 7.2 and more...

      Painter3000 started Aug 2, 2026 in Show and tell
      @Painter3000
      @Painter3000
      0
    • You must be logged in to vote
      ๐Ÿ’ฌ

      [Research] LLM GRPO Training on RX 7900 XTX โ€” 3 Actionable Findings for RDNA3 Optimization

      ChenHongYu2026 started Jul 30, 2026 in General
      @ChenHongYu2026
      @ChenHongYu2026
      0
    • You must be logged in to vote
      ๐Ÿ’ฌ

      Strix Halo (gfx1151) - 3x speedup on Hunyuan Video with 7.10 + HSA_USE_SVM=0 fix

      bkpaine1 started Dec 22, 2025 in General
      @bkpaine1 @wszgrcy
      @bkpaine1 @wszgrcy
      1
    • You must be logged in to vote
      ๐Ÿ’ก

      ROCmPorter: open-source tool that scans repos for CUDA lock-in and opens ROCm/HIP migration PRs

      pavansai20052004-hue started Jul 22, 2026 in Ideas
      @pavansai20052004-hue
      @pavansai20052004-hue
      1
    You are visitor 01402520 โ€” GeezerVison does not store the pages you look up.
    Best viewed with Netscape Navigator 4.7 or Internet Explorer 6 at 800ร—600. No cookies required.