Skip to content

fix async continuous batching #429

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
aniketmaurya merged 4 commits into from
Feb 18, 2025
Merged

fix async continuous batching #429

Changes from all commits
Commits
Show all changes
4 commits
Select commit Hold shift + click to select a range
8f1832f
fix vllm
aniketmaurya Feb 18, 2025
3140761
fix
aniketmaurya Feb 18, 2025
9a89a73
update
aniketmaurya Feb 18, 2025
332c89f
[pre-commit.ci] auto fixes from pre-commit.com hooks
pre-commit-ci[bot] Feb 18, 2025
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
5 changes: 3 additions & 2 deletions src/litserve/loops/continuous_batching_loop.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -135,8 +135,8 @@ async def prefill(
if self.has_capacity(lit_api):
response_queue_id, uid, _, input = request
logger.debug(f"New request: {uid}, {input}")
self.add_request(uid, input, lit_api, lit_spec)
self.response_queue_ids[uid] = response_queue_id
self.add_request(uid, input, lit_api, lit_spec)
else:
response_queue_id, uid, _, input = request
pending_requests.append((response_queue_id, uid, input))
Expand Down Expand Up @@ -197,7 +197,8 @@ async def run(
# Process one step for all active sequences
responses = await self.step(prev_outputs, lit_api, lit_spec)
if len(responses) == 0:
raise HTTPException(500, "No responses from step()")
logger.warning("No responses from step() but has_active_requests() is true")
continue
if responses and not isinstance(responses[0], Output):
raise HTTPException(500, "Expected StepOutput from step()")

Expand Down
Loading