Skip to content

Vllm omni action policy #236

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
MaciejBalaNV wants to merge 2 commits into
base: main
Choose a base branch
from
+419 −1
Open

Vllm omni action policy #236

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
2 commits
Select commit Hold shift + click to select a range
6572dec
Added policy action notebook
MaciejBalaNV Jun 23, 2026
0231853
Removed websocket example
MaciejBalaNV Jun 23, 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: 5 additions & 1 deletion cookbooks/cosmos3/generator/action/README.md
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -115,7 +115,9 @@ generation (see [`run_fd_with_vllm.ipynb`](./run_fd_with_vllm.ipynb) and
| `flow_shift` | `10.0` |

The notebooks build the full request body for AV, DROID, and UMI examples,
including autoregressive chunked generation for the robotics examples.
including autoregressive chunked generation for the robotics examples. Policy
inference uses async `POST /v1/videos` to retrieve a rollout video plus
top-level `action` metadata.

### VLLM-Omni Notebook Walkthrough

Expand All @@ -126,6 +128,8 @@ write outputs under `outputs/cosmos3_action_vllm/`:
DROID, and UMI robotics examples.
- [`run_id_with_vllm.ipynb`](./run_id_with_vllm.ipynb) — inverse dynamics,
predicting ego-motion trajectories from input AV videos.
- [`run_policy_with_vllm.ipynb`](./run_policy_with_vllm.ipynb) — policy
inference for DROID through the async video API.



Expand Down
Loading