Discussions

• 1 You must be logged in to vote

  🙏

  Question on KVCache

  Bond1995 asked Oct 15, 2025 in Q&A · Unanswered

  

  

  0
• 4 You must be logged in to vote

  💬

  HIPBLASLT error, and the work around for AMD/ROCM users who are getting it

  lemon07r started Jun 21, 2024 in General

  

  

  

  

  4
• 1 You must be logged in to vote

  💡

  Graph Neural Network Library for LibTorch - Seeking Community Input

  devesht21 started Aug 27, 2025 in Ideas

  

  

  0
• 1 You must be logged in to vote

  🙏

  How do you make torchtune deterministic?

  Sinestro38 asked Jul 15, 2025 in Q&A · Unanswered

  

  

  2
• 1 You must be logged in to vote

  🙏

  How do you do LoRA on torchtune with only lora applied to MLP?

  Sinestro38 asked Jul 7, 2025 in Q&A · Closed · Answered

  

  

  6
• 2 You must be logged in to vote

  🙏

  How do you specify to use ZeRO-2, DDP, or ZeRO-3 (FSDP) in torchtune?

  Sinestro38 asked Jul 8, 2025 in Q&A · Unanswered

  

  

  1
• 1 You must be logged in to vote

  🙏

  LLaMa 3.2 vision - Clarification on User Input Masking During Training

  ronyadgar asked Jul 12, 2025 in Q&A · Unanswered

  

  

  0
• 1 You must be logged in to vote

  💬

  [bike shed] tune binary name

  vadimkantorov started May 30, 2025 in General

  

  

  2
• 1 You must be logged in to vote

  🙏

  How to train a layer before the first layer of existing models (e.g. gemma) using torchtune?

  GeorgeCarpenter asked May 28, 2025 in Q&A · Answered

  

  

  1
• 1 You must be logged in to vote

  💬

  An illegal memory access was encountered with more then 1 GPU

  mariozupan started Apr 26, 2025 in General

  

  

  5
• 1 You must be logged in to vote

  💬

  Models other than available in torchtune.models

  sovit-123 started Apr 27, 2025 in General · Closed

  

  

  4
• 1 You must be logged in to vote

  🙏

  Does Llama 3.2 vision use CLIP?

  vedantroy asked Mar 5, 2025 in Q&A · Answered

  

  

  

  

  2
• 4 You must be logged in to vote

  📣

  Welcome to torchtune discussions!

  joecummings announced May 9, 2024 in Announcements

  

  

  

  

  2
• 1 You must be logged in to vote

  🙏

  Having trouble fine tuning after upgrading to torchtune 0.5

  troy256 asked Mar 18, 2025 in Q&A · Unanswered

  

  

  2
• 1 You must be logged in to vote

  🙏

  Tensor Parallel implementation of LoRA

  TAplutos asked Mar 3, 2025 in Q&A · Unanswered

  

  

  1
• 2 You must be logged in to vote

  💡

  MlFlow Implementation

  fabiogeraci started Dec 3, 2024 in Ideas · Closed

  

  

  

  

  9
• 1 You must be logged in to vote

  🙏

  How does torchtune's offerings compare with what LitGPT provides?

  karrtikiyer asked Jan 22, 2025 in Q&A · Answered

  

  

  1
• 1 You must be logged in to vote

  🙏

  Evaluate between epochs

  aldro61 asked Dec 28, 2024 in Q&A · Closed · Unanswered

  

  

  1
• 1 You must be logged in to vote

  🙏

  Are there plans to offer QAT recipes for other configurations?

  troy256 asked Oct 30, 2024 in Q&A · Unanswered

  

  

  

  

  2
• 1 You must be logged in to vote

  🙏

  Is there a standard way to pass tools parse the tool calling messages in Llama 3.2?

  albertbou92 asked Nov 6, 2024 in Q&A · Closed · Answered

  

  

  2
• 2 You must be logged in to vote

  🙏

  Can I do SFT with dataset that includes tool usage with TorchTune?

  albertbou92 asked Oct 30, 2024 in Q&A · Answered

  

  

  6
• 2 You must be logged in to vote

  🙏

  Integrating and Fine-Tuning a Custom Llama 3.1 Model in Llama 3.2

  Hyperakan asked Oct 10, 2024 in Q&A · Answered

  

  

  

  

  3
• 1 You must be logged in to vote

  🙏

  CUDA out of memory with llama3.2-3b on A100?

  troy256 asked Oct 14, 2024 in Q&A · Closed · Unanswered

  

  

  2
• 1 You must be logged in to vote

  💬

  Next release and schedule

  oharapaGitHub started Sep 13, 2024 in General · Closed

  

  

  2
• 1 You must be logged in to vote

  🙏

  Possible to fine tune on 2 different dataset classes?

  troy256 asked Sep 6, 2024 in Q&A · Answered

  

  

  8