huggingface/optimum
🚀 Accelerate training and inference of 🤗 Transformers and 🤗 Diffusers with easy to use hardware optimization tools
PythonApache-2.0
Pinned issues
Issues
- 2
Support for `torch.export.export`
#1893 opened by kilianyp - 7
onnx export for cuda does not work
#1892 opened by geraldstanje - 2
RuntimeError: Expected all tensors to be on the same device, but found at least two devices
#1889 opened by Daya-Jin - 0
Trying to export a cohere model, that is a custom or unsupported architecture, but no custom onnx configuration was passed as `custom_onnx_configs`.
#1888 opened by yimingqian - 0
onnx optimum ORTOptimizer inference runs slower than setfit.export_onnx runtime.InferenceSession inference
#1885 opened by geraldstanje - 0
- 1
Phi-3 support for openvino export not working
#1880 opened by jojo1899 - 1
Quantizing from cli using config file fails
#1838 opened by jens-totemic - 0
- 0
Unable to generate question-answering model for Llama and there is also no list of what are the supported models for question-answering
#1876 opened by customautosys - 4
Phi3 support
#1826 opened by martinlyons - 1
Add LLava ONNX export has a problem
#1873 opened by Pengjie-W - 0
- 0
ONNX converted Whisper model takes more than twice the VRAM of the torch version
#1869 opened by Hubert-Bonisseur - 0
`Graph Output Error` when loading optimizing model after running `optimizer.optimize`
#1868 opened by Ashton-Sidhu - 1
how to change optimum temporary path ?
#1855 opened by neonarc4 - 1
- 1
circular dependency issue
#1863 opened by jamesrobertwilliams - 2
trOCR in onnx format does not read full text
#1858 opened by feff2 - 0
Improve inference time TrOCR
#1859 opened by CrasCris - 2
Issue Report: Unable to Export Gemma 7b Model to ONNX Format in Optimum
#1814 opened by Harini-Vemula-2382 - 0
trOCR in onnx format does not read full text
#1857 opened by feff2 - 0
ai21labs/Jamba-tiny-random support
#1854 opened by frankia312 - 5
Idefics2 Support in Optimum for ONNX export
#1821 opened by gtx-cyber - 1
Llama 3 Support
#1835 opened by bitterspeed - 0
[Tracking] TorchDynamo ONNX exporter issues
#1810 opened by BowenBao - 0
gemma-2b static quantized, generate text makes no sense
#1853 opened by CHNtentes - 0
no attribute '_TASKS_TO_AUTOMODELS' error
#1852 opened by klutzDrawers - 1
AttributeError: '_ORTModelForWhisper' object has no attribute '_retrieve_segment'
#1816 opened by MrRace - 0
The Whisper large-v3 model exported to ONNX does not return the end timestamp for the last chunk
#1850 opened by IlyaPikin - 0
Add support for grounding-dino model type
#1849 opened by JeroendenBoef - 0
OPT6.7b ONNX model not giving accurate results on CPU
#1848 opened by pragyam32 - 0
Static Quantization for Seq2Seq models like T5
#1847 opened by NQTri00 - 1
Can't get ORTStableDiffusionPipeline to run on GPU on neither AWS nor GCP fresh instances
#1844 opened by iuliaturc - 0
Low performance of THUDM/chatglm3-6b onnx model
#1846 opened by tuhinpahari - 0
- 0
Support for speech to text models.
#1843 opened by JamesBowerXanda - 2
Output names magic in recent optimum for onnx export
#1842 opened by jobergum - 2
- 0
why does ORTModelForCausalLM assume new input length is 1 when past_key_values is passed
#1839 opened by cyh-ustc - 0
Pushing to the hub with a token
#1836 opened by IlyasMoutawwakil - 0
The transformation of the model Blip2ForConditionalGeneration to BetterTransformer failed
#1833 opened by garyzhang99 - 2
Support Transformers >=4.40.0
#1822 opened by saattrupdan - 0
Request for ONNX Export Support for Blip Model in Optimum
#1818 opened by n9s8a - 1
- 1
Support Llava ONNX export
#1813 opened by Harini-Vemula-2382 - 4
Exporting tinyllama-1.1b using onnxruntime bf16 crashes
#1807 opened by mgiessing - 4
- 0
advice for simple onnxruntime script for ORTModelForVision2Seq (or separate encoder/decoder)
#1804 opened by eduardatmadenn - 0
Issue Report: Unable to Export Qwen Model to ONNX Format in Optimum
#1798 opened by Harini-Vemula-2382