PI-Mamba L18 Depth4 CC-XAttn LIBERO no-LoRA fullVLM

30k-step LIBERO finetuned checkpoint for the PI-Mamba context-interface ablation with token-level ctxD cross-attention and no gate.

Checkpoint

  • OpenPI config: pi05_mamba_tiny_ctxd18_crossattn_fullvlm_libero_5090
  • Experiment: icra_nolora_fullvlm_pimamba_l18_crossattn_depth4_cosine5e5_5e6_b16_acc1_seed42
  • Step: 30000
  • Initializer: pi05_base_pytorch
  • Adaptation: full VLM / full vision / full action path, no LoRA
  • VLM prefix: L18
  • Mixer: Mamba depth4
  • Context interface: CC-XAttn, token-level ctxD cross-attention
  • Training: batch 16, grad accumulation 1, seed 42
  • LR schedule: cosine from 5e-5 to 5e-6 over 30k steps
  • LIBERO eval: seed 7, 50 trials per task, 10 denoising steps

LIBERO Results

Mixer Context Spatial Object Goal LIBERO-10 Average Action expert core Train time Median latency
Mamba CC-XAttn 98.6 99.4 94.6 94.2 96.7 0.017B 13.83 h 119.66 ms

Use

--config-name pi05_mamba_tiny_ctxd18_crossattn_fullvlm_libero_5090
--checkpoint-dir /path/to/checkpoint

The checkpoint includes the LIBERO norm stats under assets/physical-intelligence/libero.

Downloads last month

-

Downloads are not tracked for this model. How to track
Safetensors
Model size
3B params
Tensor type
F32
·
BF16
·
Video Preview
loading

Dataset used to train giakhuyendihoc/pimamba-l18-depth4-crossattn-libero-nolora-fullvlm