Discussions

• 33 You must be logged in to vote

  🙌

  MLX Community Projects

  awni started Feb 8, 2024 in Show and tell

  

  

  

  

  58
• 1 You must be logged in to vote

  💬

  I'm getting strange behavior when trying to use compile.

  bug Something isn't working

  cianciosa started Oct 14, 2025 in General

  

  

  4
• 1 You must be logged in to vote

  💡

  [Feature Request] XLA (and NumPy) compatible sorting for NaNs

  polvalente started Oct 10, 2025 in Ideas

  

  

  3
• 1 You must be logged in to vote

  🙏

  Code example for stop_gradient()

  RahulBhalley asked Jan 29, 2024 in Q&A · Answered

  

  

  

  

  7
• 1 You must be logged in to vote

  🙏

  Does reshape make a copy of the data?

  louisamand asked Oct 6, 2025 in Q&A · Answered

  

  

  1
• 2 You must be logged in to vote

  💬

  Any plans on the new Metal 4 tensor API integration？

  kemchenj started Oct 1, 2025 in General

  

  

  1
• 1 You must be logged in to vote

  🙏

  Would it be possible to use SIMD for matmul?

  jli05 asked Sep 30, 2025 in Q&A · Unanswered

  

  

  0
• 1 You must be logged in to vote

  🙏

  MLX matmul (and therefore tensordot) 2,000+ times faster than Numpy on AArch64 CPU. How?

  jli05 asked Sep 30, 2025 in Q&A · Answered

  

  

  1
• 1 You must be logged in to vote

  💬

  Notes on building mlx with uv

  zcbenz started May 1, 2025 in General

  

  

  1
• 1 You must be logged in to vote

  💬

  Control flow in compile/vmap

  ghcollin started Aug 18, 2025 in General

  

  

  1
• 1 You must be logged in to vote

  🙏

  Data layout

  cianciosa asked Aug 13, 2025 in Q&A · Closed · Unanswered

  

  

  1
• 6 You must be logged in to vote

  📣

  MLX on CUDA

  awni announced Jul 25, 2025 in Announcements

  

  

  

  

  11
• 2 You must be logged in to vote

  🙏

  Converting LoRA (adapters.safetensors) to GGUF

  stippi2 asked Oct 21, 2024 in Q&A · Answered

  

  

  

  

  7
• 2 You must be logged in to vote

  🙏

  Metal Code completion

  yaroslavyaroslav asked Jul 16, 2025 in Q&A · Unanswered

  

  

  2
• 1 You must be logged in to vote

  🙏

  Documentation First or Source Code First？

  Redempt1onzzZZ asked Jul 15, 2025 in Q&A · Unanswered

  

  

  0
• 1 You must be logged in to vote

  🙏

  Selective actor, critic freezing while sharing the same context in loss function

  tianenchong asked Jul 10, 2025 in Q&A · Answered

  

  

  4
• 1 You must be logged in to vote

  🙏

  MLX on Mac as model experimentation and Nvidia CUDA deployment

  tianenchong asked Jul 6, 2025 in Q&A · Answered

  

  

  1
• 1 You must be logged in to vote

  🙏

  Write to input array in a *shader* to circumvent output array creation

  hlovatt asked Jun 27, 2025 in Q&A · Unanswered

  

  

  1
• 1 You must be logged in to vote

  🙏

  Advice on custom kernels with bit manipulations and recursion

  louisamand asked Jun 24, 2025 in Q&A · Unanswered

  

  

  2
• 1 You must be logged in to vote

  🙌

  Thanks for your framework, I used it in a demo

  tennlee started Jun 8, 2025 in Show and tell

  

  

  1
• 1 You must be logged in to vote

  💡

  Triton Integration for High-Performance CUDA Kernels?

  NripeshN started Jun 16, 2025 in Ideas

  

  

  0
• 1 You must be logged in to vote

  💡

  Loading models with mmap

  polyrand started Feb 3, 2024 in Ideas

  

  

  

  

  6
• 1 You must be logged in to vote

  💡

  Leveraging MoE Sparse activations with LoRA?

  openSourcerer9000 started Jun 6, 2025 in Ideas

  

  

  0
• 1 You must be logged in to vote

  🙏

  Is example NN Quick Start code incorrect?

  hlovatt asked May 30, 2025 in Q&A · Answered

  

  

  1
• 1 You must be logged in to vote

  🙏

  Jupyter caching shaders

  hlovatt asked May 26, 2025 in Q&A · Unanswered

  

  

  0