microsoft / Olive

Olive is an easy-to-use hardware-aware model optimization tool that composes industry-leading techniques across model compression, optimization, and compilation.

https://microsoft.github.io/Olive/

microsoft/Olive Issues

Get Error while optimizing SDXL of DirectML example
Updated 16 days ago
"num_images" doesn't work for the example of directml stable_diffusion_xl.
Updated 17 days ago
onnx
Closed 20 days ago
safetensor model
Closed 20 days ago
huggingface_hub.errors.HFValidationError: Repo id must use alphanumeric chars or '-', '_', '.', '--' and '..' are forbidden, '-' and '.'
Closed 20 days ago
Llava-7b model Conversion to ONNX and Latency Optimization - OOM error (even after setting paging file size)
Updated 25 days ago2
Vitis quantization is broken with ORT 1.18
Closed a month ago2
Enabling openai/whisper-large-v3 using olive-ai-0.6.0 [onnxruntime-gpu: 1.17.1] on Intel CPU/GPU is not supporting
Updated a month ago2
Error on the Generate an ONNX model and optimize step
Closed a month ago5
I don't have models/optimized/llama_v2 folder after I've run python llama_v2.py --optimize
Updated a month ago8
status.IsOK() was false. Tensor shape cannot contain any negative value
Updated a month ago1
Failed to run symbolic shape inference when doing LLM Optimization with DirectML
Updated a month ago8
Olive workflow for mistral model optimization does not work
Updated 2 months ago16
Exception while running SD XL: Not enough memory resources are available to complete this operation
Updated 2 months ago1
pydantic.error_wrappers.ValidationError: 7 validation errors for RunConfig
Closed 2 months ago1
[FR]: FlashAttention support for Whisper
Updated 2 months ago1
[Bug] Optimization step for unet fails after 'Protobuf parsing failed'
Closed 2 months ago10
Whisper-medium conversion failed
Closed 2 months ago8
Early stopping in Hugging Face models
Closed 2 months ago7
whisper pipeline corrupting the model, unable to run on DML EP
Closed 2 months ago1
[Bug]: Merged ONNX model exceeds 2GB, the model will not be checked without `save_path` given.
Closed 2 months ago5
GenAIModelExporter Component - parameter mismatch
Closed 2 months ago3
Whisper model does not work If you add a flag --enable_timestamps
Closed 2 months ago7
models_rank.json issue
Closed 2 months ago4
Missing dependency: psutil
Closed 2 months ago2
Whisper model converted via onnxruntime 1.17.1 won't work
Closed 3 months ago5
ImportError: cannot import name 'load_model' from 'onnxruntime.quantization.quant_utils'
Closed 5 months ago13
Any workarounds to convert phi-2 model on Windows
Closed 3 months ago3
SDXL crashing when trying to run
Closed 3 months ago3
NOT_IMPLEMENTED : Could not find an implementation for BeamSearch(1) node with name 'BeamSearch_node'
Closed 3 months ago6
This is an invalid model
Updated 3 months ago4
Conversion of some models are buggy
Updated 4 months ago
Unable to perform Whisper GPU Int8 conversion
Updated 4 months ago11
Error with search strategic.py 'Conversion Merged" has no output models for Llama2 optimization
Closed 4 months ago2
UnboundLocalError: local variable 'output_model_json' referenced before assignment
Closed 5 months ago9
Is this pass flow possible for Stable Diffusion?: OrtTransformersOptimization → IncDynamicQuantization or IncStaticQuantization
Updated 5 months ago46
Missing file https://github.com/microsoft/Olive/blob/main/docs/source/overview/configuring_pass
Closed 5 months ago2
assert conversion_footprint and optimizer_footprint with Stable Diffusion conversion
Closed 5 months ago8
Validation errors for HfConfig
Closed 5 months ago5
AssertionError - assert conversion_footprint and optimizer_footprint when converting a model with size>512
Updated 5 months ago2
Whisper does not converted using onnxruntime-directml
Updated 5 months ago2
[Bug]: stable diffusion performance degrade
Closed 5 months ago2
Converted Whisper do not way to return timestamps
Closed 6 months ago5
Unable to convert openai/whisper-large-v3
Closed 6 months ago3
Optimized model slower than original one on CUDAExecutionProvider
Updated 6 months ago12
No safety checker in SDXL
Closed 6 months ago3
[Bug]: AssertionError: No valid accelerator specified for target system.
Closed 6 months ago4
How do I create a SDXL Olive model that is optimized for various Execution providers - DirectML, CPU, GPU
Closed 6 months ago8
[Bug]: local_resource_path.save_to_dir(save_dir, save_name, overwrite) failed
Closed 6 months ago9
[FR]: Pass func_kwargs to _evaluate_custom
Closed 6 months ago1