All pull requests

Pull requests

• Use packaging instead of pkg_resources in the TensorFlow backend

  #2597·David-Wu1119 opened Sep 23, 2026·

  ·
  1

• Use the non-deprecated CPU autocast API

  #2593·Shubham-Padkonde opened Sep 21, 2026·

  ·
  2

• Multimodal benchmarks

  #2590·lkk12014402 opened Sep 18, 2026Contributor·

  ·

• fix(fp8_quant test): pass htexp_device_type to the unexpected-device error message

  #2568·Anai-Guo opened Sep 5, 2026·

  ·

• Add evaluation and serving scripts for vLLM integration

  #2559·chensuyue opened Aug 26, 2026Contributor·

  ·

• support ifbench for LLMs.

  #2542·lkk12014402 opened Aug 5, 2026Contributor·

  ·
  4

• [JAX] Add empty change to run PR tests - DO NOT MERGE

  #2500·anko-intel opened Jun 17, 2026Contributor·

  ·

• [won't merge]add debug notes

  #2465·yiliu30 opened May 9, 2026Contributor·

  ·

• Add evaluation support for DS V4

  #2463·yiliu30 opened May 7, 2026Contributor·

  ·

• Add batch size and gradient accumulation parameters to quantization s…

  #2456·xin3he opened Apr 28, 2026Contributor·

  ·
  1

• Fix activation scale inf issue for const_weight and const_scale

  #2448·qgao007 opened Apr 15, 2026Contributor·

  ·
  8

• Add mark_step() for the sliced FusedSDPA

  #2395·yangulei opened Jan 23, 2026·

  ·
  4

• Split fp8_fused_sdpa into two phases

  #2346·czhu15 opened Nov 26, 2025·

  ·
  8

• [WIP] Add mlperf example for whisper

  #2343·lkk12014402 opened Nov 25, 2025Contributor·

  ·

• [Draft] Enable slicing of fp8 FusedSDPA for APC

  #2340·yangulei opened Nov 20, 2025·

  ·

• [WIP] Add mlperf examples

  #2338·lkk12014402 opened Nov 18, 2025Contributor·

  ·

• Add a feature of UniformQDQ - support CV/NLP model's OPs, includingConv, DepthwiseConv2D, MatMul, etc. Additional op support be added upon request.

  #2155·qgao007 opened Mar 27, 2025Contributor·

  ·
  16