Pi05-Base-CoreAI / README.md
kevinqz's picture
privacy: scrub host chip/OS from model card
e7b4f39 verified
|
Raw
History Blame Contribute Delete
7.54 kB
---
license: gemma
base_model: lerobot/pi05_base
pipeline_tag: robotics
library_name: coreai
extra_gated_heading: "Access this Gemma Model Derivative"
extra_gated_prompt: >-
These weights are a Model Derivative of Gemma. Access and use are governed by the
Gemma Terms of Use (https://ai.google.dev/gemma/terms) and the Gemma Prohibited
Use Policy (https://ai.google.dev/gemma/prohibited_use_policy). By requesting access
you agree to those terms and the Section 3.2 use restrictions, and you agree to carry
these terms forward on any further distribution.
extra_gated_fields:
I agree to the Gemma Terms of Use and the Gemma Prohibited Use Policy: checkbox
I agree to carry these terms forward on any redistribution: checkbox
tags:
- coreai
- core-ai
- coreai-fabric
- aimodel
- coreml
- apple
- apple-silicon
- on-device
- robotics
- vla
- vision-language-action
- action
---
> **Canonical:** [`kevinqz/Pi05-Base-CoreAI`](https://huggingface.co/kevinqz/Pi05-Base-CoreAI) β€” source of truth.
# Pi0.5 base Gemma (fabric)
> ⚠️ **Robot policy β€” needs a matching robot to actuate.** This is
> [lerobot/pi05_base](https://huggingface.co/lerobot/pi05_base) converted to an
> Apple Core AI `.aimodel`. Its output is a **normalized action chunk** for
> **generalist_multi_robot**. Run it on any other robot, or with mismatched calibration /
> normalization stats, and it emits floats that *look* valid but **actuate
> garbage** β€” the most dangerous silent-failure mode. It is a **base checkpoint**:
> fine-tune on your robot's data before expecting useful behavior.
An Apple Core AI conversion of
[lerobot/pi05_base](https://huggingface.co/lerobot/pi05_base) β€” a **robot policy**
that maps images + proprioceptive state (+ a language instruction, when the model
uses one) to a continuous action chunk (flow_matching sampler). Produced by
[coreai-fabric](https://github.com/kevinqz/coreai-fabric/blob/main/recipes/pi05-base.yaml) and indexed by
[coreai-catalog](https://github.com/kevinqz/coreai-catalog).
## Model facts
| Field | Value |
|---|---|
| Parameters | 3.3B |
| Architecture | transformer |
| Capabilities | vision-language-action, robotics |
| Embodiment | generalist_multi_robot |
| Sampling | flow_matching (10-step) |
| Quantization / precision | none / float16 |
| On-disk size | 6.0 GB |
| Asset kind | split-export policy (encode, denoise_step) + norm_stats |
| assetVersion | 2.0 |
## Use it β€” this needs host code you supply
A policy is **not** a chat model: there is no stock high-level Swift runtime for
it. The bundle is the split-export shape β€” an `encode` graph (run once per
observation) + a `denoise_step` graph (the host drives it `num_steps` times) +
`norm_stats.json` (un-normalization). **You supply the host loop** (the N-step
sampler + un-normalization) in Swift. Recommended integration: keep LeRobot's
Python `RobotClient` for the servos/cameras/calibration, and run inference
on-device β€” see the `io_contract` in the catalog for the exact tensors.
```bash
pip install coreai-catalog && coreai-catalog install pi05-base
```
## Requirements
- **Deployment: macOS 27.0+ / iOS 27.0+, Xcode 27+.** The asset serializes with `minimum_os v27`,
so the on-device Swift runtime requires macOS/iOS 27+. A Mac on macOS 26 can
convert and inspect it but not run it on-device.
- Apple Silicon. A matching robot to actuate (see the banner).
## Verification (action parity)
- **Gate A (structure): passed** β€” the bundle's layout + metadata were
validated on real hardware; the graphs load.
- **Gate B β€” action_parity: 96.2% min chunk-cosine** vs the fp16 reference over 16 frames of `recorded episodes`, 10-step, fixed-noise (measured on Apple Silicon).
- **Near-zero-action note (full disclosure):** the 96.2% is the raw MIN over all frames, dominated by frame(s) where the reference action is ~zero (the robot barely moves) and chunk-cosine is mathematically ill-conditioned. The **median chunk-cosine is 99.9%** and the **max ABSOLUTE per-dim action error is 0.0193** (normalized) β€” i.e. behaviorally faithful everywhere. This model class is gated on the absolute action error (behaviorally meaningful) since cosine is undefined for ~zero actions; every raw number is reported here and in `parity-report.json`. fabric never fakes a number.
- This certifies the export is **numerically faithful to the source policy** β€” it
does **NOT** certify real-world task success, embodiment transfer, or
closed-loop stability. Closed-loop success-rate eval (LIBERO/ManiSkill):
**not_run** (a separate future gate). Reproduce with `coreai-fabric verify`.
## Provenance
| Field | Value |
|---|---|
| Base model | [lerobot/pi05_base](https://huggingface.co/lerobot/pi05_base) @ `7de663972b7817d2c4cf2d84c821153dfea772e9` |
| Converted by | `models/pi05/export.py` (version not reported) |
| Recipe | [pi05-base](https://github.com/kevinqz/coreai-fabric/blob/main/recipes/pi05-base.yaml) (recipe_source: fabric) |
| Precision / quantization | float16 / none |
| Conversion date | 2026-07-05 |
Machine-readable, in this repo:
[`parity-report.json`](./parity-report.json) (gate results) Β·
[`reproduce-manifest.json`](./reproduce-manifest.json) Β· [`LICENSE`](./LICENSE)
(upstream terms).
## License and attribution
Weights Β© Google LLC and the upstream authors, licensed **gemma** β€” see the bundled `LICENSE`. This artifact is a **converted derivative** of the base policy: its
weights were converted to Apple Core AI format. The conversion itself is
community work.
These weights are a **Model Derivative of Gemma** β€” the upstream pi0/pi0.5 policy embeds PaliGemma/Gemma parameters. Redistribution and use are governed by the [Gemma Terms of Use](https://ai.google.dev/gemma/terms) (bundled as `LICENSE`) and restricted by the [Gemma Prohibited Use Policy](https://ai.google.dev/gemma/prohibited_use_policy) (Β§3.2, incorporated by reference). Access is gated: you must accept these terms before downloading, and must carry them forward β€” including the mandated NOTICE and the Β§3.2 restrictions β€” on any further distribution (see [`NOTICE`](./NOTICE)). **Gemma** and **PaliGemma** are trademarks of Google LLC, used here only descriptively; this conversion is **not affiliated with, produced by, or endorsed by Google**.
## Links
- **Base model:** [lerobot/pi05_base](https://huggingface.co/lerobot/pi05_base)
- **Reproduce:** [recipe `pi05-base`](https://github.com/kevinqz/coreai-fabric/blob/main/recipes/pi05-base.yaml)
- **Index:** [coreai-catalog](https://github.com/kevinqz/coreai-catalog) β€” the
neutral registry tying upstream ↔ this asset ↔ mirror together
- [HF Collection](https://huggingface.co/collections/kevinqz/coreai-apple-on-device-6a4879f21c7e1a87c99bcf5a)
## The on-device Core AI ecosystem
- [coreai-fabric](https://github.com/kevinqz/coreai-fabric) β€” the reproducible
recipe β†’ `.aimodel` pipeline that produced this asset.
- [coreai-catalog](https://github.com/kevinqz/coreai-catalog) β€” the index of Core
AI models across the community, with provenance and integration snippets.
- [apple/coreai-models](https://github.com/apple/coreai-models) β€” Apple's official
exporters and runtimes.
- [LeRobot](https://huggingface.co/lerobot) β€” the upstream robotics ecosystem.
## Not affiliated with Apple
Community conversion. Not produced, hosted, or endorsed by Apple. Apple and Core
AI are trademarks of Apple Inc., used here only to describe the target
runtime/format. This is an independent community conversion.