Skip to content

fix: pin verl<0.7.0 to prevent incompatible upgrade breaking vLLM async server #499

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

Draft
Copilot wants to merge 4 commits into
base: main
Choose a base branch
from
Draft

fix: pin verl<0.7.0 to prevent incompatible upgrade breaking vLLM async server #499

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
4 commits
Select commit Hold shift + click to select a range
a855e37
Initial plan
Copilot Feb 28, 2026
19c72db
fix: handle empty tensors in compute_data_metrics to prevent crash on…
Copilot Feb 28, 2026
7d56275
chore: remove accidentally committed cache files and add to .gitignore
Copilot Feb 28, 2026
7690d1e
fix: pin verl<0.7.0 to prevent incompatible upgrade breaking vLLM asy…
Copilot Feb 28, 2026
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: 4 additions & 2 deletions pyproject.toml
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -34,7 +34,7 @@ apo = [
# It's not recommended to use agentlightning[verl] to install VERL and its dependencies.
# though it's listed here for completeness.
verl = [
"verl>=0.5.0",
"verl>=0.5.0,<0.7.0",
"vllm>=0.8.4", # Due to interface change of ExternalZeroMQDistributedExecutor
]

Expand Down Expand Up @@ -140,7 +140,9 @@ torch-gpu-stable = [
{include-group = "torch-cu128"},
"flash-attn>=2.8.3",
"tensordict>=0.9.1",
"verl>=0.6.0",
# verl 0.7.0 has breaking API changes (SPMD rollout removal, ResourcePool changes)
# that are incompatible with the current codebase
"verl>=0.6.0,<0.7.0",
]
# Use this instead of --group torch-legacy --group torch-gpu
torch-gpu-legacy = [
Expand Down
Loading