Starred repositories
An autonomous agent that profiles, optimizes, and validates AI training code.
[NeurIPS 2026] Beyond Data Scaling: Representation-Centric Continued Pre-training for Vision-Language-Action Models
RPent: Agentic Infrastructure for the Physical World
[CVPR 2026 Highlight] LitePT: Lighter Yet Stronger Point Transformer
Argus is an open source dataset-and-format-agnostic harness for the densest annotation of robot learning data.
A Survey of Reinforcement Learning for Large Reasoning Models
A curated list of reinforcement learning with verifiable rewards (continually updated)
Rethinking Visual-Language-Action Model Scaling: Alignment, Mixture, and Regularization
[ICRA 2026] VITRA: Scalable Vision-Language-Action Model Pretraining for Robotic Manipulation with Real-Life Human Activity Videos
Official repository for "AM-RADIO: Reduce All Domains Into One"
A Data Streaming Library for Efficient Neural Network Training
An all-in-one VLA engineering platform for embodied AI — from data to real-robot deployment.
Official JAX implementation of End-to-End Test-Time Training for Long Context
Cross-Embodiment Video World Models are Zero-Shot Physical Simulators
Flex-π: A multi-stream world-action model with compute flexibility: one checkpoint that deploys as a VLA, a full world model, or anything in between.
From Human Videos to Robot Manipulation: A Survey on Scalable Vision-Language-Action Learning with Human-Centric Data
18
Updated Jun 2, 2026
Implementation of the paper "Beyond Action Residuals: Real-World Robot Policy Steering via Bottleneck Latent Reinforcement Learning" (https://arxiv.org/abs/2605.19919).
This repo is the official implementation of "τ0-VLA: a Hierarchical Robot Fo 6CA5 undation Model with World-Model-Guided Test-Time Computation".
Open Lakehouse Format for Multimodal AI. Convert from Parquet in 2 lines of code for 100x faster random access, vector index, and data versioning. Compatible with Pandas, DuckDB, Polars, Pyarrow, a…
verl/HybridFlow: A Flexible and Efficient RL Post-Training Framework
Open-source speech recognition toolkit for training, inference, streaming ASR, VAD, punctuation, speaker diarization pipelines, and OpenAI-compatible/MCP serving.
DeepSeek Harness: Everything is a Plugin.
RoboBrain 2.5: Advanced version of RoboBrain. Depth in Sight, Time in Mind. 🎉🎉🎉