You can deploy this model and start earning money today!

sancov/
snapflow_ctpf08-test

Name: snapflow_ctpf08-test
Availability: InStock

sancov/smolvla_recap_libero_spatial

apache-2.0

Views

No views yet

sancov/snapflow_ctpf08-test — SmolVLA+RECAP on LIBERO (1-step SnapFlow suffix, CFG weight 1.0)

Reference checkpoint for TensorRT conversion accuracy benchmarking in ros-fhnw-autonomy (docker/benchmark/). It is published so the on-device benchmark job can load real trained weights: half precision saturates on the outlier activation channels that only appear in trained weights, so an FP16 engine checked against a randomly initialised policy passes a check it would fail in deployment.

Policy: smolvla_recap (1-step SnapFlow suffix, CFG weight 1.0)
Validated against: sancov/smolvla_recap_libero_spatial
Inputs: observation.images.image, observation.images.image2 (3x256x256), observation.state (8)
Output: action (7), chunk size 20

SnapFlow-distilled: distinct weights, not the 10-step model with a flag flipped.

Used as the eager reference in the eager-vs-TensorRT comparison; not intended as a general-purpose LIBERO policy release.