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

Openai api supports lora path #2247

Closed
wants to merge 6 commits into from
Closed

Openai api supports lora path #2247

Changes from all commits
Commits
Show all changes
6 commits
Select commit Hold shift + click to select a range
67e8c97
Update llama.py
ccchow Oct 21, 2024
fdb6ca2
Merge branch 'sgl-project:main' into sage-attention
ccchow Oct 21, 2024
490264e
Merge pull request #1 from ccchow/sage-attention
ccchow Oct 21, 2024
724c28e
Merge branch 'sgl-project:main' into main
ccchow Oct 29, 2024
5d85f24
Merge branch 'sgl-project:main' into main
ccchow Nov 13, 2024
dbd401c
Merge branch 'sgl-project:main' into openai-api-supports-lora_path
ccchow Nov 28, 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
6 changes: 5 additions & 1 deletion python/sglang/srt/models/llama.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -45,6 +45,7 @@
from sglang.srt.model_executor.forward_batch_info import ForwardBatch
from sglang.srt.utils import make_layers

from sageattention import sageattn

class LlamaMLP(nn.Module):
def __init__(
Expand Down Expand Up @@ -167,7 +168,10 @@ def forward(
qkv, _ = self.qkv_proj(hidden_states)
q, k, v = qkv.split([self.q_size, self.kv_size, self.kv_size], dim=-1)
q, k = self.rotary_emb(positions, q, k)
attn_output = self.attn(q, k, v, forward_batch)

# attn_output = self.attn(q, k, v, forward_batch)
attn_output = sageattn(q, k, v, is_causal=False, smooth_k=True)

output, _ = self.o_proj(attn_output)
return output

Expand Down