| ---
|
| license: apache-2.0
|
| base_model: XiaomiRobotics/Xiaomi-Robotics-1-RoboCasa365
|
| language:
|
| - en
|
| - zh
|
| tags:
|
| - robotics
|
| - vision-language-action
|
| - robocasa365
|
| - originx
|
| ---
|
|
|
| # OriginX
|
|
|
| OriginX is a robot policy for RoboCasa365, developed by Qinzhen Ma using [Xiaomi-Robotics-1-RoboCasa365](https://huggingface.co/XiaomiRobotics/Xiaomi-Robotics-1-RoboCasa365) as its foundation.
|
|
|
| [Project website](https://robocasa.originxairobotics.com/) · [Code and inference guide](https://github.com/Quinn-Ma/OriginX) · [Technical report](https://github.com/Quinn-Ma/OriginX/blob/main/TECHNICAL_REPORT.md)
|
|
|
| ## Complete model weights
|
|
|
| This repository contains the complete inference weight set: the three full base-model shards, tokenizer and processor assets, the **A2000** Action LoRA adapter, and the **B2000** continuous conditioning branch. The base-model weights are included here; a separate download from another model repository is not required.
|
|
|
| | Component | Files |
|
| |---|---|
|
| | Full frozen base | `model-00001-of-00003.safetensors`, `model-00002-of-00003.safetensors`, `model-00003-of-00003.safetensors` |
|
| | A2000 Action LoRA | `adapter-originx-2000.pt` |
|
| | B2000 conditioning branch | `branch-00002000.pt` |
|
| | Model assets | Configuration, weight index, tokenizer, and processor files |
|
|
|
| The base has 5,053,149,696 parameters stored in BF16. Its three shards total 10,106,433,336 bytes. The adaptation files are the unchanged weights used in the recorded evaluation; they remain separate to preserve the original loading and numerical behavior.
|
|
|
| **Release naming.** A2000 is the public name of the existing adapter. It completed 1,613 optimizer updates over 80,000 sampled windows; B2000 completed 2,000 updates over 256,000 sampled windows. Renaming does not alter weight bytes, training history, or evaluation results.
|
|
|
| ## Run the model
|
|
|
| Implementation code lives on [GitHub](https://github.com/Quinn-Ma/OriginX). Clone that repository and follow [PORTABLE_INFERENCE.md](https://github.com/Quinn-Ma/OriginX/blob/main/PORTABLE_INFERENCE.md). The `prepare_originx.py` helper downloads this model snapshot, verifies the original asset hashes, and combines its weights with the three matching model implementation files from GitHub. It prepares `assets/base` and `assets/weights` for the loader. No Python source files are hosted in this Hugging Face repository.
|
|
|
| The frozen upstream revision is `3a6d0293bfa90759d34a7fc48c2c62413cd7bcf4`. `SHA256SUMS` covers the published model assets. The publication loader has CPU contract checks; a new real-model GPU parity run through that helper has not been performed.
|
|
|
| ## Evaluation
|
|
|
| The complete author-run native-reset evaluation recorded **1,496/2,500 successes (59.84%)** across 50 tasks, with 50 episodes per task. All 1,004 policy failures are retained, with zero missing or infrastructure-unknown outcomes.
|
|
|
| | Split | Successes / episodes | Rate |
|
| |---|---:|---:|
|
| | Atomic-Seen | 747 / 900 | 83.00% |
|
| | Composite-Seen | 479 / 800 | 59.875% |
|
| | Composite-Unseen | 270 / 800 | 33.75% |
|
|
|
| [RoboCasa submission #26](https://github.com/robocasa-benchmark/leaderboard/pull/26) is pending organizer review. An earlier separate 600-pair comparison found a 1.00 percentage-point difference with reported 95% interval [-2.17, 4.17] and McNemar p=0.6173; it did not establish a reliable gain. The full evaluation has no matched full baseline or branch ablation. The [technical report](https://github.com/Quinn-Ma/OriginX/blob/main/TECHNICAL_REPORT.md) retains the protocol, training lineage, and limitations; [evaluation evidence](https://github.com/Quinn-Ma/OriginX/releases/tag/v1.0.0) remains on GitHub.
|
|
|
| ## License
|
|
|
| Apache-2.0. See `LICENSE` and `NOTICE` for attribution. Source code, reports, and evaluation archives are maintained on GitHub; this repository distributes model weights and their configuration/documentation.
|
|
|
