MODEL / CHECKPOINT
MimicLite 031b8d67
RoboParty
Model
Checkpoint artifacts and an independent MotionChallenge inference adapter.
- Organization
- RoboParty
- Type
- Dense
Dense MLP policy with normalized hidden features. - Version
- SHA 031b8d67
checkpoint SHA256 - Robot support
- Unitree G1 · 29 DoF
- Author source revision
- 3253810ce21dbcc05d8f4942a017ea3eaffc6cf1
Checkpoint artifacts
- Author release
- EGalahad/mimic-lite ↗
- Artifact type
- Author-released checkpoint
- Original file
- PPO / policy.onnx ↗
The author README links this PPO Drive folder, whose recorded listing contains the exact policy file ID used by the adapter. The download host is Google Drive, not a separate model publisher.
Download links deliver the adapter files. The release links and verification notes distinguish author files from conversions; SHA256 identifies the bytes used.
Download ↗
031b8d67557795d86e4ed3f30e7a707414534feba65f330c64f1b07a1f5c7a0bInference interface
The adapter declares its control clock and reference window below. Source and configuration describe the observation and action transforms.
Observation requirements
Model-specific handling
031b8d67 is the ONNX SHA256 prefix, not a W&B run ID. Reference translation is centered on the REFERENCE anchor; the heading transform uses permitted orientation. No actual-position estimate.
Exact checkpoint & evidence ↗- Control interval
- 20 ms
- Reference history
- 160 ms
- Reference preview
- 80 ms
- Output interface
- joint-pd-feedforward
- Configuration
- policy.json