All pull requests

Pull requests

• Upgrade python to 3.11 for all workflows

  #2026·echarlaix opened Sep 24, 2026Collaborator·

  ·
  1

• Scope GITHUB_TOKEN permissions per job

  #2025·hf-security-analysis[bot] opened Sep 24, 2026Bot·

  ·
  1

• Add LTX video 0.9.5 version support

  #2024·JohnLeFeng opened Sep 24, 2026·

  ·
  1

• [Qwen3-ASR] Ignore RoPE Pattern For Qwen3 ASR Model

  #2014·anzr299 opened Sep 21, 2026Contributor·

  ·
  2

• [OpenVINO] Support PaddleOCR-VL-1.5 with task image-text-to-text

  #2013·openvino-agent opened Sep 21, 2026Contributor·

  ·

• Bump huggingface/doc-builder/.github/workflows/upload_pr_documentation.yml from cf20b09f274383f0d91a7055c5e43f0a2ab3d1a3 to 9f9bb430a712b5650d0e43e510656098ecdd75d4 in the actions group

  dependenciesgithub_actions
  #2012·dependabot[bot] opened Sep 21, 2026Bot·

  ·
  1

• [OpenVINO] Support LFM2.5-VL-3B with task image-text-to-text

  #2009·openvino-agent opened Sep 18, 2026Contributor·

  ·

• Add OpenVINO export support for spark2_5 architecture (Spark-X2.5)

  #2008·bharqav opened Sep 18, 2026·

  ·
  1

  3

• [OpenVINO] Support LFM2.5-VL-3B with task image-text-to-text

  #2005·popovaan-bot opened Sep 17, 2026·

  ·
  1

• Add slow T5 tokenizer support

  #2002·JohnLeFeng opened Sep 17, 2026·

  ·
  1

• Exclude models that were not correctly saved in the cache

  #2001·echarlaix opened Sep 16, 2026Collaborator·

  ·
  2

• Make modeling_visual_language importable on newer transformers (VLLM 0.29 support)

  #1994·3QL opened Sep 12, 2026·

  ·
  2

• Improve OpenVINO export errors for unwritable output filesystems

  #1993·Gunjan00001 opened Sep 11, 2026·

  ·
  7

• Fix of MoE fusion for Granite-MoE.

  #1992·popovaan opened Sep 11, 2026Collaborator·

  ·
  1

• [OpenVINO] Support academic-ds-9B with task text-generation

  #1990·openvino-agent opened Sep 10, 2026Contributor·

  ·

• [OpenVINO] Support Unlimited-OCR with task image-text-to-text

  #1987·openvino-agent opened Sep 10, 2026Contributor·

  ·

• Add ministral3 (Ministral-3-3B-Reasoning-2512) OpenVINO export support

  #1986·AshutoshSinghIntel opened Sep 10, 2026Contributor·

  ·
  21

• [OpemVINO] Qwen3-ASR implement vlm like audio_encoder + text_embedding + language model export

  #1985·as-suvorov opened Sep 10, 2026Collaborator·

  ·
  7

• [OpenVINO] Add DFlash 2 two-model export

  #1984·imargulis opened Sep 10, 2026·

  ·

• Support Muse-Glimmer Dflash method

  #1983·openvino-agent opened Sep 9, 2026Contributor·

  ·

• Add OpenVINO GenAI AIPC test workflow and report

  #1980·helena-intel opened Sep 7, 2026Collaborator·

  ·
  29

• [EXPERIMENT][WIP][OpenVINO] Add support for LFM2-VL

  #1979·mlukasze opened Sep 5, 2026Contributor·

  ·
  2

• [EXPERIMENT][WIP][OpenVINO] Add support for baidu/Unlimited-OCR

  #1978·mlukasze opened Sep 5, 2026Contributor·

  ·

• [OpenVINO] Support Dflash 2 for Qwen3.8-27B

  #1972·rkazants opened Sep 3, 2026Collaborator·

  ·
  1

• Don't make optional dependencies hard requirements of Eagle3/DFlash draft-model export

  #1969·blairducrayoppat opened Sep 3, 2026·

  ·
  1