Issues · huggingface/optimum-intel · GitHub

Labels Milestones

Add OpenVINO export support for FLUX.2-dev / Flux2Pipeline

#1767

· MichaelDevine opened

on Jun 5, 2026

[OpenVINO] Support Gemma 4 12B with model type gemma4_unified

#1764

· OpenArcBob opened

on Jun 5, 2026

Native OpenVINO export for Qwen3.6 / Qwen3.5 MoE (model_type=qwen3_5_moe)

#1754

· bob798 opened

on May 28, 2026

Native support for DeepSeek-V4 (model_type=deepseek_v4)

#1748

· bob798 opened

on May 26, 2026

Bug: Qwen3.5-9B INT4 export produces incoherent output (hybrid linear/full attention model)

#1722

· daviburg opened

on May 6, 2026

Qwen3.6 (qwen3_5 / qwen3_5_moe) text-generation export blocked — only image-text-to-text task registered

#1721

· bobvious opened

on May 5, 2026

[Bug] Mixtral 8x22B (and other large MoE) cannot be exported on a sub-400 GB RAM machine — full model is loaded in one shot

#1708

· t8 opened

on Apr 30, 2026

[Bug] Misleading "Unable to read the model" error when intermediate IR is truncated during INT4 export

#1707

· t8 opened

on Apr 30, 2026

[Code Quality] Replace Python runtime asserts with ValueError in ERNIE-Image model patchers

#1698

· Rishabh-Sahni-0809 opened

on Apr 22, 2026

[Bug] Platform-specific code and hardcoded paths in test_ernie_image.py break tests on Windows

#1697

· Rishabh-Sahni-0809 opened

on Apr 22, 2026

Bug: sentence-transformers export fails due to removed _text_length attribute in version 5.4.0

#1680

· MissLostCodes opened

on Apr 13, 2026

Remote Code Execution via LlavaQwen2OpenVINOConfig and Phi3VisionOpenVINOConfig

#1668

· Vancir opened

on Mar 31, 2026