Welcome to the official model repository for nanoVLA, a high-frequency, lightweight Vision-Language-Action (VLA) suite optimized for bimanual manipulation on the I2RT Bimanual YAM (Dual 7-DoF Arm) platform with 3 cameras (Overhead, Left Wrist, Right Wrist).
All checkpoints below are packaged and downloadable directly from this repository:
| Model Checkpoint | Architecture & Key Innovation | 14-DoF Action MSE | Frequency / Latency | Primary Strength |
|---|---|---|---|---|
spatial_chunked_flagship.pt |
588 Spatial Patches + Motion Cross-Attn + H=16 Chunking | 0.19648 | 30.6 Hz | 🥇 Flagship Multi-Camera Policy |
arena_champion.pt |
Continuous Smooth L1 Huber + 14-DoF Proprioception | 0.00689 | 64.0 Hz | 🥇 Lowest Single-Step Action MSE |
foveated_champion.pt |
Foveated Dynamic Spatial Token Pruning (66% Pruned) | 0.14443 | 73.1 Hz ⚡ | 2.4x Speedup + Visual Salience |
dual_speed_champion.pt |
Dual-Speed Fast/Slow Hierarchy (5 Hz / 1000 Hz) | 0.03567 | 907.7 Hz 🚀 | 1.1 ms High-Frequency Impedance |
action_chunking_champion.pt |
H=16 Action Chunking + Temporal Ensembling | 0.22064 | 49.1 Hz | 0.027 rad/s² Silky Smooth Jitter |
so400m_champion.pt |
Scaled SigLIP-SO400M-384 Backbone (900.1M Params) | 0.06013 | 4.3 Hz | Rich Zero-Shot Spatial Grounding |
critic_recovery_champion.pt |
Verification Critic & Anomaly Recovery Detector | 0.12926 | >10,000 Hz | Real-time Slip & Grasp Monitoring |
bimanual_coupled_champion.pt |
Cross-Arm Attention & Relative Distance Constraint | 0.16939 | 250.9 Hz | Prevents Dropping Grasped Objects |
fk_consistent_champion.pt |
Differentiable Forward Kinematics 3D Metric Loss | 0.13258 | 50.0 Hz | 3D Millimeter Cartesian Precision |
rvq_hybrid_champion.pt |
Residual Vector Quantization (K=64 Codebook) | 0.11779 | 50.3 Hz | Multi-Modal + Sub-mm Precision |
generalization_champion.pt |
Robust Domain Randomization (5-Strata Invariant) | 0.03944 | 64.0 Hz | Extreme Lighting / Sensor Noise |
Prompt: "bimanual pick and place red cube"
│
[Top Camera] [Left Wrist Cam] [Right Wrist Cam] [SigLIP Text Tokens]
(3, 224, 224) (3, 224, 224) (3, 224, 224) (1, 16)
│ │ │ │
▼ ▼ ▼ ▼
┌──────────────┐ ┌──────────────┐ ┌──────────────┐ ┌──────────────┐
│ SigLIP-Base │ │ SigLIP-Base │ │ SigLIP-Base │ │ Language │
│ Vision (86M) │ │ Vision (86M) │ │ Vision (86M) │ │ Embedding │
└──────┬───────┘ └──────┬───────┘ └──────┬───────┘ └──────┬───────┘
│ 196 tokens │ 196 tokens │ 196 tokens │ 16 tokens
└──────────────────────┴──────────┬────────┴──────────────────────┘
│ 588 + 16 = 604 tokens
▼
┌──────────────────────────────┐
│ nanoVLA Transformer Trunk │
│ (6 Layers, 384 Dim, 6 Heads)│
└──────────────┬───────────────┘
│
▼
┌────────────────────┐
│ 14-DoF Action Head │
│ Left (7) + Right(7)
└────────────────────┘
import torch
from model import NanoVLA, NanoVLAConfig
# Load Flagship Champion Checkpoint
ckpt = torch.load("spatial_chunked_flagship.pt", map_location="cuda")
model = NanoVLA(ckpt['config']).cuda()
model.load_state_dict(ckpt['model_state_dict'])
model.eval()
# Predict 14-DoF action for 3 cameras
imgs = torch.rand(1, 3, 3, 224, 224, device="cuda")
txt = torch.zeros(1, 16, dtype=torch.long, device="cuda")
proprio = torch.zeros(1, 14, device="cuda")
with torch.no_grad():
actions = model.predict_action(imgs, txt, proprio_state=proprio)
print("Predicted 14-DoF Actions (7 Left + 7 Right):", actions.cpu().numpy())
MIT License (c) 2026 npow. Developed for fast, accessible, and robust robot learning.
so400m_champion.pt 3.6 GBhifi_robust_champion.pt 1.2 GBbimanual_coupled_champion.pt 866 MBspatial_chunked_flagship.pt 866 MBtemporal_motion_champion.pt 865 MBnanodva_best.pt 865 MBflow_chunk_champion.pt 864 MBflow_matching_champion.pt 863 MBmae_spatial_champion.pt 863 MBaction_chunking_champion.pt 862 MBgrand_master_champion.pt 862 MBkeypoint_ik_champion.pt 862 MBcontrastive_invariance_champion.pt 862 MBcritic_recovery_champion.pt 862 MBdynchunk_champion.pt 862 MBfoveated_champion.pt 861 MBrvq_hybrid_champion.pt 861 MBgeneralization_champion.pt 861 MBcollision_avoidance_champion.pt 861 MBfk_consistent_champion.pt 861 MBarena_champion.pt 861 MBbest_ckpt.pt 861 MBckpt.pt 861 MBapex_champion.pt 860 MBrope_3d_champion.pt 860 MBdual_speed_champion.pt 847 MB