Issues

• [Proposal] Add SmolLM2 lightweight on-device model example

  ml-explore/mlx-examples#1456 · Filter by author Vedag812 opened on Oct 1, 2026

• segment_anything: fuse attention with mx.fast.scaled_dot_product_attention

  ml-explore/mlx-examples#1451 · Filter by author bahaehmimdi opened on Sep 7, 2026

• Community project: Depth-Anything-V2 running via a from-scratch PyTorch-API shim (torch-mlx)

  ml-explore/mlx-examples#1449 · Filter by author bahaehmimdi opened on Sep 7, 2026

• Proposal: neural surface-code decoder (QEC) example?

  ml-explore/mlx-examples#1445 · Filter by author rajanshxrma opened on Aug 26, 2026

• Suggestion: add a Stable Fast 3D (SF3D) example — native MLX single-image-to-3D-mesh

  ml-explore/mlx-examples#1443 · Filter by author bahaehmimdi opened on Aug 22, 2026

• [whisper] mlx_whisper CLI overwrites output file when multiple audio inputs are provided

  ml-explore/mlx-examples#1441 · Filter by author bryzgaloff opened on Aug 18, 2026

• whisper: numerical parity tests fail with MLX 0.32.0 on Apple Silicon

  ml-explore/mlx-examples#1440 · Filter by author sfeng01-tech opened on Aug 18, 2026

• mlx_lm.fuse can not work

  ml-explore/mlx-examples#1436 · Filter by author williwms opened on Aug 6, 2026

• Stable Diffusion: Transformer2D's GroupNorm uses MLX's default eps (1e-5) where diffusers uses 1e-6

  ml-explore/mlx-examples#1434 · Filter by author andreinknv opened on Jul 31, 2026

• [Proposal] DeepSeek-OCR example — SAM + CLIP dual vision encoder → DeepSeek-V2 MoE

  ml-explore/mlx-examples#1433 · Filter by author vignesh-kumar-v opened on Jul 30, 2026

• whisper: align no_speech_threshold fallback condition with openai/whisper

  ml-explore/mlx-examples#1427 · Filter by author maciej opened on Jun 18, 2026

• whisper/convert.py outputs model.safetensors but mlx_whisper.load_models expects weights.safetensors

  ml-explore/mlx-examples#1421 · Filter by author kinoshitakashi01 opened on May 1, 2026
  1

• [whisper] KeyError: 'words' on transcription (with --task translate)

  ml-explore/mlx-examples#1418 · Filter by author uogbuji opened on Apr 24, 2026

• [whisper] Add Flash Attention and batched decoding for up to 10x speedup

  ml-explore/mlx-examples#1412 · Filter by author ilyasmukiev opened on Mar 26, 2026
  3

• [whisper] Feature Request: Add carry_initial_prompt parameter to transcribe()

  ml-explore/mlx-examples#1410 · Filter by author malleroid opened on Mar 13, 2026

• [Model Request] Porting Nvidia PersonaPlex-7B (Moshi architecture) & Mimi Codec

  ml-explore/mlx-examples#1405 · Filter by author Vlor999 opened on Feb 12, 2026
  1

• Question: (mlx_whisper) How to eliminate spurious "you" transcriptions

  ml-explore/mlx-examples#1402 · Filter by author Napoleon-BlownApart opened on Dec 24, 2025
  1

• Using fine tuned whisper models in MLX must support beam search

  ml-explore/mlx-examples#1391 · Filter by author FredrikKarlssonSpeech opened on Nov 6, 2025

• How to fine tune quantized modeling convert it to GGUF still been quantized?

  ml-explore/mlx-examples#1382 · Filter by author kudla opened on Aug 14, 2025

• Support for Qwen-Image generation

  ml-explore/mlx-examples#1379 · Filter by author psm-2 opened on Aug 4, 2025
  2

• Convert method for only for pytorch/safetensors to MLX

  ml-explore/mlx-examples#1374 · Filter by author debasisdwivedy opened on Jul 21, 2025

• Is there a way to implement a Timeout / cancellation for transcribe() method, espeicially model.decode()? (large-v3 sometimes stalls >200s on ~10s audio)

  ml-explore/mlx-examples#1373 · Filter by author esphoenixc opened on Jul 18, 2025
  1

• Segfault during inference

  ml-explore/mlx-examples#1368 · Filter by author maxlund opened on Jun 4, 2025
  4

• mlx-whisper OOM error on files > 1GB

  ml-explore/mlx-examples#1366 · Filter by author stickystyle opened on May 5, 2025
  4

• text to video model support?

  ml-explore/mlx-examples#1335 · Filter by author JohnSmithyy opened on Mar 11, 2025