Issues

• Feature request: export AutoRound-quantized checkpoints to OpenVINO

  huggingface/optimum-intel#2052 · Filter by author blairducrayoppat opened on Oct 6, 2026

• Error while exporting Qwen/Qwen-Image-2.1 to OpenVINO

  huggingface/optimum-intel#2050 · Filter by author azhuvath opened on Oct 6, 2026

• [Guidance] multiple image edit using Qwen image 2.1

  huggingface/optimum-intel#2023 · Filter by author ak-sangal opened on Sep 24, 2026

• Add OpenVINO export support for spark2_5 architecture (Spark-X2.5-4B / 1.7B)

  huggingface/optimum-intel#2003 · Filter by author spf1983 opened on Sep 17, 2026
  1

• [OpenVINO] Eagle3 tooling: greedy-equality test can't detect a degraded drafter; export needs einops, declared only in TESTS_REQUIRE

  huggingface/optimum-intel#1964 · Filter by author blairducrayoppat opened on Aug 29, 2026
  3

• OpenVINO export can silently produce an IR that cannot use PagedAttention (no fused SDPA op)

  huggingface/optimum-intel#1948 · Filter by author aweussom opened on Aug 21, 2026
  

• Feature Request: Qwen-Image-Edit OpenVINO Support

  huggingface/optimum-intel#1934 · Filter by author ak-sangal opened on Aug 15, 2026
  9

• Feature Request] Support phi4-siglip / Phi-4-reasoning-vision-15B architecture in OpenVINO export

  huggingface/optimum-intel#1921 · Filter by author jonfon96 opened on Aug 8, 2026

• [OpenVINO] Support Speculative Decoding in Gemma 4 with model type 'gemma4_assistant' and 'gemma4_unified_assistant'

  huggingface/optimum-intel#1878 · Filter by author whatevermarch opened on Jul 19, 2026

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

  huggingface/optimum-intel#1767 · Filter by author MichaelDevine opened on Jun 5, 2026

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

  huggingface/optimum-intel#1764 · Filter by author OpenArcBob opened on Jun 5, 2026
  7

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

  huggingface/optimum-intel#1754 · Filter by author bob798 opened on May 28, 2026
  1

• Native support for DeepSeek-V4 (model_type=deepseek_v4)

  huggingface/optimum-intel#1748 · Filter by author bob798 opened on May 26, 2026

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

  huggingface/optimum-intel#1722 · Filter by author daviburg opened on May 6, 2026
  6

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

  huggingface/optimum-intel#1721 · Filter by author bobvious opened on May 5, 2026
  7

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

  huggingface/optimum-intel#1708 · Filter by author t8 opened on Apr 30, 2026

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

  huggingface/optimum-intel#1707 · Filter by author t8 opened on Apr 30, 2026

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

  huggingface/optimum-intel#1698 · Filter by author Rishabh-Sahni-0809 opened on Apr 22, 2026

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

  huggingface/optimum-intel#1697 · Filter by author Rishabh-Sahni-0809 opened on Apr 22, 2026

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

  huggingface/optimum-intel#1680 · Filter by author MissLostCodes opened on Apr 13, 2026

• Remote Code Execution via LlavaQwen2OpenVINOConfig and Phi3VisionOpenVINOConfig

  huggingface/optimum-intel#1668 · Filter by author Vancir opened on Mar 31, 2026
  4

• Feature Request: Native OpenVINO export and NNCF quantization support for SeamlessM4T-v2 (seamless_m4t_v2)

  huggingface/optimum-intel#1667 · Filter by author seferdemirci opened on Mar 30, 2026

• Add support for PaddleOCR-VL-1.5

  huggingface/optimum-intel#1654 · Filter by author purofle opened on Mar 27, 2026

• Add support for moonshotai/Kimi-K2.5

  huggingface/optimum-intel#1647 · Filter by author MissLostCodes opened on Mar 24, 2026
  1

• Add tests which check, that required transformations are applied

  good first issue
  huggingface/optimum-intel#1645 · Filter by author popovaan opened on Mar 23, 2026
  4