All pull requests

Pull requests

• Qwen3-MoE: fused-expert quantisation, BF16 router and expert-wise GPTQ

  #324·Shreyas8612 opened Sep 17, 2026Collaborator·

  ·
  3

• Phase-split (prefill/decode) quantisation for Llama and Qwen3

  #323·Shreyas8612 opened Sep 17, 2026Collaborator·

  ·
  7

• Fix minifloat RMSNorm output dtype, add E3M4 MXFP format, drop state_dict deepcopy

  #322·Shreyas8612 opened Sep 17, 2026Collaborator·

  ·
  3

• Fix MX block quantization: row-local blocking, OCP shared exponent, element rounding

  #321·Shreyas8612 opened Aug 7, 2026Collaborator·

  ·
  8

• Llama phase split quantisation support

  #320·YuxuanHan0326 opened May 21, 2026Collaborator·

  ·

• MX PTQ quantization pass updates, tutorial, and PLENA version release

  #318·jiaeenie opened May 12, 2026Collaborator·

  ·

• Submission for Group 18

  #317·idrees-mahmood opened Mar 27, 2026·

  ·

• Group 1 - Hardware-Aware Transformer Optimisation: Integrating Programmable Attention, Triton Kernel Fusion, and Multi-Objective NAS

  #315·aahaidar01 opened Mar 27, 2026·

  ·

• Adapting MASE for Edge Deployment (Group 05 2026 Project Submission) MR2

  #314·Souparna-Mandal opened Mar 27, 2026·

  ·

• fix: resolve TypeError in BlockMinifloatQuantize backward pass

  #302·Shreyas8612 opened Feb 4, 2026Collaborator·

  ·
  1

• Refactoring Progress

  documentationrefactor
  #292·Aaron-Zhao123 opened Nov 1, 2025Collaborator·25 of 93 tasks·

  ·
  

• Verified & Integrated MXINT Implementation and Evaluation on NeRF (ADLS Group 0)

  #277·omaralkhatib03 opened Mar 27, 2025·

  ·

• Feature/add snip pruning

  #274·amagri16 opened Mar 23, 2025·

  ·

• Fix Lab4 Hardware Emit

  #271·firemountain154B opened Mar 13, 2025Contributor·

  ·

• feat: integrate spikeLM

  #266·JeffreyWong20 opened Feb 18, 2025Collaborator·

  ·

• Upgrading to python3.13? [testing needed]

  #252·Aaron-Zhao123 opened Feb 3, 2025Collaborator·

  ·
  3

  

• onn transform pass

  #250·Johnny1882 opened Jan 23, 2025Contributor·

  ·

• merge mxint_linear

  #229·firemountain154B opened Oct 23, 2024Contributor·

  ·
  8

• support autogenerating transformer block in vision transformer

  #225·firemountain154B opened Aug 28, 2024Contributor·

  ·

• Research/alpa light

  #224·pgimenes opened Aug 9, 2024Collaborator·

  ·

• Added init latency printing so we can observe and reduce test cases

  #212·jianyicheng opened Jul 15, 2024Collaborator·

  ·

• Added insert fork pass

  #187·jianyicheng opened May 17, 2024Collaborator·

  ·

• Feature/bert

  #186·pgimenes opened May 15, 2024Collaborator·

  ·

• Mlp example

  #172·jianyicheng opened Apr 24, 2024Collaborator·

  ·
  17