Skip to content

Fix Qwen3.5-0.8B WebGPU optimization config #388

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
xiaofeihan1 wants to merge 6 commits into
base: main
Choose a base branch
from
Open

Fix Qwen3.5-0.8B WebGPU optimization config #388

Show file tree
Hide file tree
Changes from 1 commit
Commits
Show all changes
6 commits
Select commit Hold shift + click to select a range
21add70
Fix Qwen3.5-0.8B WebGPU optimization config
xiaofeihan1 Apr 29, 2026
2e427f0
fix
xiaofeihan1 Apr 30, 2026
52cef83
Merge branch 'main' into fix/qwen3.5-webgpu-config
apsonawane May 4, 2026
d66c314
Merge branch 'main' into fix/qwen3.5-webgpu-config
apsonawane May 6, 2026
775f3bf
Merge branch 'main' into fix/qwen3.5-webgpu-config
apsonawane May 7, 2026
f721a26
fix session_options
xiaofeihan1 May 8, 2026
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
2 changes: 1 addition & 1 deletion Qwen-Qwen3.5-0.8B/builtin/optimize.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -55,7 +55,7 @@ def update_genai_config(output_dir: str = MODELS_DIR, device: str = "cpu"):

vision_session_options = {"log_id": "onnxruntime-genai", "provider_options": vision_provider_options}

config["model"]["decoder"]["session_options"] = session_options
config["model"]["decoder"]["session_options"] = vision_session_options
Comment thread
xiaofeihan1 marked this conversation as resolved.
Outdated

config["model"]["embedding"] = {
"filename": "embedding.onnx",
Expand Down
12 changes: 12 additions & 0 deletions Qwen-Qwen3.5-0.8B/builtin/webgpu/embedding.json
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -7,6 +7,17 @@
"io_config": "get_embedding_io_config",
"dummy_inputs_func": "get_embedding_dummy_inputs"
},
"systems": {
"local_system": {
"type": "LocalSystem",
"accelerators": [
{
"device": "gpu",
"execution_providers": ["WebGpuExecutionProvider"]
}
]
}
},
"passes": {
"convert": {
"type": "OnnxConversion",
Expand Down Expand Up @@ -40,6 +51,7 @@
"external_data_name": "embedding.onnx.data"
}
},
"target": "local_system",
"no_artifacts": true,
"output_dir": "webgpu/models/embedding.onnx"
}
12 changes: 12 additions & 0 deletions Qwen-Qwen3.5-0.8B/builtin/webgpu/text.json
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -3,6 +3,17 @@
"type": "HfModel",
"model_path": "Qwen/Qwen3.5-0.8B"
},
"systems": {
"local_system": {
"type": "LocalSystem",
"accelerators": [
{
"device": "gpu",
"execution_providers": ["WebGpuExecutionProvider"]
}
]
}
},
"passes": {
"m": {
"type": "ModelBuilder",
Expand All @@ -16,5 +27,6 @@
}
},
"no_artifacts": true,
"target": "local_system",
"output_dir": "webgpu/models/text.onnx"
}
12 changes: 12 additions & 0 deletions Qwen-Qwen3.5-0.8B/builtin/webgpu/vision.json
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -7,6 +7,17 @@
"io_config": "get_vision_io_config",
"dummy_inputs_func": "get_vision_dummy_inputs"
},
"systems": {
"local_system": {
"type": "LocalSystem",
"accelerators": [
{
"device": "gpu",
"execution_providers": ["WebGpuExecutionProvider"]
}
]
}
},
"passes": {
"c": {
"type": "OnnxConversion",
Expand Down Expand Up @@ -71,6 +82,7 @@
"external_data_name": "vision.onnx.data"
}
},
"target": "local_system",
"no_artifacts": true,
"output_dir": "webgpu/models/vision.onnx"
}
Loading