Skip to content

fix(collector): align model dtype for vLLM FlashInfer fp8 KV attention #1382

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
kaim-eng wants to merge 1 commit into
base: main
Choose a base branch
from
Open

fix(collector): align model dtype for vLLM FlashInfer fp8 KV attention #1382

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
4 changes: 4 additions & 0 deletions collector/vllm/collect_attn.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -207,6 +207,10 @@ def run_attention_torch(

# Mock flashinfer's get_per_layer_parameters if needed
if backend_name_str == "FLASHINFER":
if use_fp8_kv_cache:
# FlashInferMetadataBuilder asserts kv_cache_spec.dtype ==
# model_config.dtype; KV spec uses fp8 while model_config stays bf16.
vllm_config.model_config.dtype = kv_cache_spec.dtype
import unittest.mock

from vllm.v1.attention.backends.utils import PerLayerParameters
Expand Down
Loading