octo-small-so101-multi-task-gguf
Octo-Small as vla.simd GGUF for CPU inference. Each file is self-contained.
| File | Converted from |
|---|---|
octo-small-so101-multi-task.gguf | khanhnd61/octo-small_so101-multi-task-clean |
Serve
Install vla.simd with the serve extra, then:
hf download khanhnd61/octo-small-so101-multi-task-gguf --local-dir octo-small-so101-multi-task-gguf
OMP_NUM_THREADS=$CORES vla-simd-serve --model octo \
--model-dir octo-small-so101-multi-task-gguf/octo-small-so101-multi-task.gguf --port 8080 \
--cams front,wrist \
--task "<instruction>"
Rollout
The robot side is lerobot's async client, lerobot-vla-simd:
pip install 'lerobot[async,feetech] @ git+https://github.com/khanhnd61-vr/lerobot@4b33b84296c0880ebce778d69f16a38d33825575'
On the robot, with the arm and both cameras attached:
lerobot-vla-simd --server_address=127.0.0.1:8080 --policy_type=octo \
--robot.type=so101_follower --robot.port=/dev/ttyACM0 --robot.id=my_arm \
--robot.cameras="{ front: {type: opencv, index_or_path: 0, width: 640, height: 480, fps: 30}, wrist: {type: opencv, index_or_path: 2, width: 640, height: 480, fps: 30} }" \
--actions_per_chunk=4 --chunk_size_threshold=1.0 \
--task="<instruction>"
Octo sees consecutive frames only when every frame reaches the server, hence --chunk_size_threshold=1.0.