huggingface/optimum-intel v2.1.0 on GitHub

New Model Support

Transformers v5.1-5.5 support (#1684 by @echarlaix)
Fix Kokoro voice conversion from local model directory (#1803 by @mmikolajcz)
Extended OpenVINO transformation verification tests for MoE architectures (#1776 by @anatyrova)
Unpatch 16-bit models previously patched with patch_16bit_model after conversion (#1811 by @anatyrova)
Set dynamic sequence length for FLUX.2 (#1846 by @rkazants)
Deprecated the broken contextual calibration dataset, replaced with textvqa (#1849 by @echarlaix)
Fixed passing of token_type_ids for Gemma3 (#1881 by @popovaan)
Fixed export for encoder-decoder models with unnamed tensors (#1857 by @echarlaix)
Fixed long audio inference for Qwen3-ASR (#1782 by @popovaan)
Fixed ssm states handling for Mamba and Falcon-Mamba (#1779 by @echarlaix)
Pinned huggingface_hub<1.22 to fix offline loading (#1855 by @echarlaix)
Use base transformers version for version comparisons (#1798 by @AlexanderDokuchaev)

CI: local model caching and rate-limit fixes for fork PRs (#1817, #1826, #1835, #1837, #1840, #1842 by @echarlaix)
Faster doc builds (#1851 by @mishig25)

Compatible with transformers>=4.51,<5.6

NPU-specific requirements:

Maximum supported transformers version by model architectures can be found in the table below (models not listed support up to transformers v5.6)

Details