Downloads · 30 days
0
giakhuyendihoc/pimamba-l12-depth4-libero-nolora-fullvlm
pimamba-l12-depth4-libero-nolora-fullvlm is a robotics model from giakhuyendihoc. Use it for the robotics task on the model card, and read the license before you ship it in a product.
30k-step LIBERO finetuned checkpoint for PI-Mamba L12 depth4 no-LoRA fullVLM.
Downloads · 30 days
0
Access
Public
Updated Jul 15, 2026
Parameters
3B
5.9 GB on disk
Likes
0
Public
Click a slice to open those files.
.safetensors5.9 GB · 100%
How the weights are stored.
BF163B · 100%
From the Hugging Face model README
30k-step LIBERO finetuned checkpoint for PI-Mamba L12 depth4 no-LoRA fullVLM.
pi05_mamba_tiny_early_ctxd12_gated_fullvlm_libero_5090icra_nolora_fullvlm_pimamba_l12_depth4_cosine5e5_5e6_b16_acc1_seed4230000pi05_base_pytorch5e-5 to 5e-6 over 30k stepsFiles included:
model.safetensorsmetadata.ptassets/physical-intelligence/libero/norm_stats.json| Spatial | Object | Goal | LIBERO-10 | Average |
|---|---|---|---|---|
| 98.0 | 99.8 | 98.0 | 94.0 | 97.5 |
| Metric | Value |
|---|---|
| Action expert core | 29.74M |
| Train time | 10.81 h |
| Best train step time | 1.04 s/step |
| GFLOPs/query | 3199.13 |
| Prefix + ctxD | 41.15 ms |
| Denoise | 56.06 ms |
| Denoise per step | 5.61 ms |
| Median model latency | 107.54 ms |
| Query Hz | 9.30 |
| Action Hz | 46.50 |
Use with the matching PI-Mamba/OpenPI code and config:
--config-name pi05_mamba_tiny_early_ctxd12_gated_fullvlm_libero_5090
--checkpoint-dir /path/to/checkpoint
The checkpoint includes the LIBERO norm stats under assets/physical-intelligence/libero.