Skip to content

Update text-to-image-genai notebook for FP32 precision on GPU #3348

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Open
szd5654125 wants to merge 1 commit into
base: latest
Choose a base branch
from
Open

Update text-to-image-genai notebook for FP32 precision on GPU #3348

Changes from all commits
Commits
Show all changes
1 commit
Select commit
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
9 changes: 7 additions & 2 deletions notebooks/flux.1-image-generation/flux.1-image-generation.ipynb
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -403,7 +403,8 @@
"id": "cab6790e",
"metadata": {},
"source": [
"`openvino_genai.Text2ImagePipeline` represents inference pipeline for text-to-image generation. For creation pipeline instance, you should provide directory with converted to OpenVINO model and inference device."
"`openvino_genai.Text2ImagePipeline` represents inference pipeline for text-to-image generation. For creation pipeline instance, you should provide directory with converted to OpenVINO model and inference device.\n",
"> **Note**: When using GPU, the default inference precision is FP16, which can cause image quality degradation (blurry or noisy images) with quantized models."
]
},
{
Expand All @@ -425,7 +426,11 @@
"\n",
"model_dir = model_base_dir / \"INT4\" if use_quantized_models.value else model_base_dir / \"FP16\"\n",
"\n",
"ov_pipe = ov_genai.Text2ImagePipeline(model_dir, device=device.value)"
"if \"GPU\" in device.value:\n",
" ov_pipe = ov_genai.Text2ImagePipeline(model_dir)\n",
" ov_pipe.compile(device.value, INFERENCE_PRECISION_HINT=\"f32\")\n",
"else:\n",
" ov_pipe = ov_genai.Text2ImagePipeline(model_dir, device=device.value)"
]
},
{
Expand Down