Downloads · 30 days
0
musclemimic-checkpoints-20260903
yangfy0627/musclemimic-checkpoints-20260903
musclemimic-checkpoints-20260903 is a reinforcement learning model from yangfy0627. Use it for the reinforcement learning task on the model card, and read the license before you ship it in a product.
All 24 independent Stage-1 runs are complete: 15 baseline + 9 anchor-v2. The repository also retains one historical duplicate backup (25 checkpoint directories).
Access
Public
Updated Oct 8, 2026
Repo size
11.2 GB
Likes
1
Public
Repo makeup
Click a slice to open those files.
Other11.2 GB · 100%
At a glance
- Task
- Reinforcement Learning
- Access
- Public
- Created
- Sep 5, 2026
- Updated
- Oct 8, 2026
- SHA
- 1c56e185
Model card
From the Hugging Face model README
Completed checkpoints — 2026-10-08
All 24 independent Stage-1 runs are complete: 15 baseline + 9 anchor-v2. The repository also retains one historical duplicate backup (25 checkpoint directories).
Each final checkpoint is update 39063 / 800,010,240 steps. All 344 checkpoint files were verified against the remote repository. Earlier selection and upload records below are retained as history.
MuscleMimic 80/20 selected checkpoints
Exactly six complete checkpoints from the 2026-09-03 offline snapshot: latest T0-T4 seed0 and latest T1 seed1. Later checkpoints on live training servers are outside this snapshot.
| Variant | Checkpoint | Saved at (UTC+8) |
|---|---|---|
| seed0/T0 | checkpoint_33550 | 2026-09-03T13:11:47+08:00 |
| seed0/T1 | checkpoint_39063 | 2026-08-27T15:48:11+08:00 |
| seed0/T2 | checkpoint_39063 | 2026-08-27T13:31:52+08:00 |
| seed0/T3 | checkpoint_39063 | 2026-08-27T15:00:16+08:00 |
| seed0/T4 | checkpoint_34770 | 2026-09-03T14:07:05+08:00 |
| seed1/T1 | checkpoint_33550 | 2026-09-03T13:06:58+08:00 |
Directories contain the original complete Orbax checkpoint (training state and associated metadata). Preserve the full directory when downloading. SHA256SUMS and FILE_MANIFEST.json cover every checkpoint payload. SELECTION.json records exact provenance. No old checkpoint series, W&B history, or evaluation videos are included.
hf download yangfy0627/musclemimic-checkpoints-20260903 --local-dir ./musclemimic-checkpoints
- Task
- Reinforcement Learning
- Created
- Sep 5, 2026
- Updated
- Oct 8, 2026