Skip to content

small fix for llmcompressor dense example #77

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

Merged
merged 1 commit into from
Oct 8, 2025
Merged

small fix for llmcompressor dense example #77

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
34 changes: 14 additions & 20 deletions hf_torchao_vllm/quantize_hf_model_with_llm_compressor.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -23,6 +23,18 @@ def run(
print(model)
tokenizer = AutoTokenizer.from_pretrained(model_name)

ignore_list = ["lm_head"]
if model_name == "Qwen1.5-MoE-A2.7B":
ignore_list.extend(
[
"re:.*mlp.gate$",
"re:.*mlp.shared_expert_gate$",
# also skip attention and shared expert, to focus on MoE for now
"re:.*self_attn.*",
"re:.*shared_expert.*",
]
)

if quant_type == "fp8":
# Configure the quantization algorithm and scheme.
# In this case, we:
Expand All @@ -31,16 +43,7 @@ def run(
recipe = QuantizationModifier(
targets="Linear",
scheme="FP8_DYNAMIC",
ignore=[
"lm_head",
# for Qwen MoE, but ok to just hardcode here for now
# https://github.com/vllm-project/llm-compressor/blob/33ef5f497a9801893764c6a2c880cb1f560067fa/examples/quantizing_moe/qwen_example.py#L10
"re:.*mlp.gate$",
"re:.*mlp.shared_expert_gate$",
# also skip attention and shared expert, to focus on MoE for now
"re:.*self_attn.*",
"re:.*shared_expert.*",
],
ignore=ignore_list,
)

# Apply quantization.
Expand Down Expand Up @@ -89,16 +92,7 @@ def tokenize(sample):
recipe = QuantizationModifier(
targets="Linear",
scheme="NVFP4",
ignore=[
"lm_head",
# for Qwen MoE, but ok to just hardcode here for now
# https://github.com/vllm-project/llm-compressor/blob/33ef5f497a9801893764c6a2c880cb1f560067fa/examples/quantizing_moe/qwen_example.py#L10
"re:.*mlp.gate$",
"re:.*mlp.shared_expert_gate$",
# also skip attention and shared expert, to focus on MoE for now
"re:.*self_attn.*",
"re:.*shared_expert.*",
],
ignore=ignore_list,
)

# Apply quantization.
Expand Down