Skip to content
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

fix BitNet integration for vLLM #139

Merged
merged 5 commits into from
Aug 9, 2024
Merged

fix BitNet integration for vLLM #139

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
5 commits
Select commit Hold shift + click to select a range
471a10c
fix BitNet integration for vLLM
xysmlx Aug 9, 2024
7164521
update ckpt name of BitNet integration for vLLM
xysmlx Aug 9, 2024
5eae019
format code
xysmlx Aug 9, 2024
99dce06
fix BitNet integration for vLLM native version
xysmlx Aug 9, 2024
0bdae66
merge main
xysmlx Aug 9, 2024
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
4 changes: 3 additions & 1 deletion integration/BitNet/vllm_workspace/inference_with_compress_format.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -35,7 +35,9 @@
ckpt_path,
dtype="half",
quantization="bitblas",
enforce_eager=True, # set False to enable cuda graph
# set enforce_eager = False to enable cuda graph
# set enforce_eager = True to disable cuda graph
enforce_eager=False,
) as bitnet_model:
bitbnet_outputs = bitnet_model.generate_greedy(["Hi, tell me about microsoft?"],
max_tokens=1024)
Expand Down
7 changes: 5 additions & 2 deletions integration/BitNet/vllm_workspace/inference_with_native_format.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -18,7 +18,7 @@
# get the path of the current file
current_file_path = os.path.realpath(__file__)
current_dir = os.path.dirname(current_file_path)
ckpt_path = os.path.join(current_dir, "../models/ckpt_bitnet_b1_58-3B_bitblas")
ckpt_path = os.path.join(current_dir, "../models/ckpt_bitnet_b1_58-3B")

parser = argparse.ArgumentParser(description="Inference with BitNet")
parser.add_argument(
Expand All @@ -35,8 +35,11 @@
with VllmRunner(
ckpt_path,
dtype="half",
quantization="bitnet",
quantization="bitnet_bitblas",
gpu_memory_utilization=0.5,
# set enforce_eager = False to enable cuda graph
# set enforce_eager = True to disable cuda graph
enforce_eager=False,
) as bitnet_model:
bitbnet_outputs = bitnet_model.generate_greedy(["Hi, tell me about microsoft?"], max_tokens=128)
print("bitnet inference output:")
Expand Down
Loading