Skip to content

Qualcomm: make --use_fp16 and the fp16 compiler spec agree #21867

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
psiddh wants to merge 2 commits into
base: main
Choose a base branch
from
+9 −5
Open

Qualcomm: make --use_fp16 and the fp16 compiler spec agree #21867

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
2 commits
Select commit Hold shift + click to select a range
29c9332
Qualcomm: make --use_fp16 and the fp16 compiler spec agree
psiddh Aug 12, 2026
d2ef854
Merge branch 'main' into main
psiddh Aug 15, 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 backends/qualcomm/export_utils.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -558,7 +558,7 @@ def build_executorch_binary(
),
QnnExecuTorchBackendType.kGpuBackend: generate_gpu_compiler_spec(),
QnnExecuTorchBackendType.kHtpBackend: generate_htp_compiler_spec(
use_fp16=False if quant_dtype is not None else True,
use_fp16=quant_dtype is None and custom_quantizer is None,
htp_performance_mode=qnn_config.htp_performance_mode,
),
Comment thread
psiddh marked this conversation as resolved.
}[qnn_config.backend]
Expand Down
6 changes: 4 additions & 2 deletions examples/qualcomm/custom_op/custom_ops_1.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -142,10 +142,12 @@ def main(args):
]

# Quantization
quant_dtype = QuantDtype.use_8a8w
# quant_dtype must be None for fp16, otherwise build_executorch_binary
# still lowers as a quantized graph and applies a default quantizer.
if args.use_fp16:
quantizer = None
quant_dtype, quantizer = None, None
else:
quant_dtype = QuantDtype.use_8a8w
quant_cfg = get_ptq_per_channel_quant_config()
custom_quant_annotator = CustomOpsQuantAnnotator()
custom_quant_annotator.register_annotation(
Expand Down
6 changes: 4 additions & 2 deletions examples/qualcomm/custom_op/custom_ops_2.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -150,10 +150,12 @@ def main(args):
]

# Quantization
quant_dtype = QuantDtype.use_8a8w
# quant_dtype must be None for fp16, otherwise build_executorch_binary
# still lowers as a quantized graph and applies a default quantizer.
if args.use_fp16:
quantizer = None
quant_dtype, quantizer = None, None
else:
quant_dtype = QuantDtype.use_8a8w
quant_cfg = get_ptq_per_channel_quant_config()
custom_quant_annotator = CustomOpsQuantAnnotator()
custom_quant_annotator.register_annotation(
Expand Down
Loading