Skip to content

[Bugfix] DeepSeek Accuracy #14476

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
Mar 8, 2025
Merged

[Bugfix] DeepSeek Accuracy #14476

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
12 changes: 7 additions & 5 deletions vllm/v1/attention/backends/mla/common.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -222,8 +222,7 @@
Fp8LinearGenericOp, current_platform_fp8_dtype, is_fp8)
from vllm.model_executor.layers.quantization.utils.quant_utils import (
scaled_quantize)
from vllm.model_executor.layers.rotary_embedding import (
DeepseekScalingRotaryEmbedding, RotaryEmbedding)
from vllm.model_executor.layers.rotary_embedding import RotaryEmbedding
from vllm.utils import cdiv, round_down

try:
Expand Down Expand Up @@ -626,9 +625,12 @@ def __init__(
self.qk_head_dim = qk_head_dim
self.v_head_dim = v_head_dim

self.rotary_emb = rotary_emb
self.use_yarn_rope = isinstance(rotary_emb,
DeepseekScalingRotaryEmbedding)
# Hack for V1 for now to avoid torch library overhead (since we are
# already inside an attention custom op), pull out the forward
# method from the rotary embedding and call it directly
# TODO(lucas): we should probably find a cleaner way to do this
self.rotary_emb = rotary_emb._forward_method

self.q_proj = q_proj
self.kv_b_proj = kv_b_proj
self.o_proj = o_proj
Expand Down