π₀.₅ (Pi05) is a Vision-Language-Action model from Physical Intelligence designed for open-world generalization: it evolves π₀ to generalize to entirely new environments and situations that were never seen during training. The LeRobot implementation is adapted from their open-source OpenPI repository. This policy has been trained and pushed to the Hub using LeRobot. Learn how to train and run it in the LeRobot pi05 guide, or browse the full documentation. The policy consumes these observation features and produces these action features. Inputs Outputs New to LeRobot? These guides cover the full workflow: - Install LeRobot — set up the lerobot package. - Hardware setup — assemble, wire, and…
Open-weight model · Robotics
pi05_telearms_v4_fullft_lr7.07e-5
by Helen Wang helen9975/pi05_telearms_v4_fullft_lr7.07e-5
pi05_telearms_v4_fullft_lr7.07e-5 is an open-weight model for robotics from Helen Wang. It has 4.1B parameters. At 16-bit it needs about 9.9 GB of GPU memory, which fits on 1x MI300X from $1.85 an hour, at the lowest prices in the SAVRN Index. It draws 12 downloads a month.
π₀.₅ (lerobot/pi05base) full fine-tune on mliu48/telearms-yam-cube-v4 (53,832 bimanual YAM demonstrations, 48 tasks, 30 Hz, 3 × 224×224 cameras). LeRobot 0.6.1 @ e40b58a8 with the repo's patch (tools/installlerobotpatch.py).
Parameters4.1B
Context—
Weights9.4 GB
License—
AccessOpen weights
Monthly Downloads12
Runs On
What it takes to serve pi05_telearms_v4_fullft_lr7.07e-5 (4.1B parameters): the memory its weights need at each precision, and the cheapest way to rent enough data-center GPUs to hold them.
| Precision | Weights | Memory needed | Cheapest setup | Per hour | Also fits |
|---|---|---|---|---|---|
| 16-bit | 8.3 GB | 9.9 GB | 1x MI300X (192 GB) Vultr |
$1.85 | 1x H100 $1.99 · 1x MI325X $2.00 |
| 8-bit | 4.1 GB | 5.0 GB | 1x MI300X (192 GB) Vultr |
$1.85 | 1x H100 $1.99 · 1x MI325X $2.00 |
| 4-bit | 2.1 GB | 2.5 GB | 1x MI300X (192 GB) Vultr |
$1.85 | 1x H100 $1.99 · 1x MI325X $2.00 |
Memory is the weights at that precision plus 20% for the runtime and a short context; a long context needs more. Prices are the lowest on-demand hourly rates in the SAVRN Index, read Oct 18, 2026.
pi05_telearms_v4_fullft_lr7.07e-5 on every accelerator the SAVRN Index prices, at every precision
Model Card
π₀.₅ (lerobot/pi05base) full fine-tune on mliu48/telearms-yam-cube-v4 (53,832 bimanual YAM demonstrations, 48 tasks, 30 Hz, 3 × 224×224 cameras). LeRobot 0.6.1 @ e40b58a8 with the repo's patch (tools/installlerobotpatch.py). - Full fine-tune (LLM + vision + action expert), bf16, gradient checkpointing. - 16 × H100 (2 nodes × 8), batch 16/GPU (global 256), 99,486 steps (one selected-data epoch). - Peak LR 7.07e-5, cosine decay to 7.07e-06, 1,000 warmup steps; relative actions (absolute grippers). Final checkpoint (step 99,486), training/evaluatepi.py suite settings in the RoboEval YAM simulator: 10 episodes per task, seeds 900000001–900000010, 1,800-step limit, 25-step action chunks, success…
Excerpt from the card by Helen Wang.
Identity and Version
- Repository
- helen9975/pi05_telearms_v4_fullft_lr7.07e-5
- Publisher
- Helen Wang
- Task
- Robotics
- Modality
- Control
- Library
- lerobot
- Parameters
- 4.1B parameters
- Languages
- Not stated by the source
- Revision
- 4a598baa0ecab581327eb77550820fdbc2b16341
- First published
- 2026-10-05
- Last updated
- 2026-10-08
Files and Weights
10 files, 9.4 GB in total. The weights are 3 files totalling 9.4 GB in safetensors.
Weights3 files · 9.4 GB
Configuration5 files · 1.0 MB
Documentation1 file · 1.8 KB
Repository1 file · 1.5 KB
Every file
| File | Type | Size | SHA-256 |
|---|---|---|---|
| model.safetensors | Weights | 9.4 GB | c2588ead93a0 |
| policy_postprocessor_step_0_unnormalizer_processor.safetensors | Weights | 14.9 KB | 593f9acbfc25 |
| policy_preprocessor_step_4_normalizer_processor.safetensors | Weights | 14.9 KB | 593f9acbfc25 |
| config.json | Configuration | 3.1 KB | — |
| policy_postprocessor.json | Configuration | 779 B | — |
| policy_preprocessor.json | Configuration | 3.4 KB | — |
| telearms_recipe.json | Configuration | 696 B | — |
| train_config.json | Configuration | 1.0 MB | — |
| README.md | Documentation | 1.8 KB | — |
| .gitattributes | Repository | 1.5 KB | — |
License and Download
- License
- Not stated by the source
- Access
- Open weights, no gate
- Download size
- 9.4 GB
Download from Helen Wang
Released by Helen Wang through its official repository on Hugging Face.
Memory Requirements
| Precision | Weights in memory |
|---|---|
| As published | 9.4 GB |
| 16-bit | 8.3 GB |
| 8-bit | 4.1 GB |
| 4-bit | 2.1 GB |
Weights only, from the published parameter count; the key-value cache and runtime add to this.
Questions About pi05_telearms_v4_fullft_lr7.07e-5
How much GPU memory does pi05_telearms_v4_fullft_lr7.07e-5 need?
About 9.9 GB at 16-bit and 2.5 GB at 4-bit: the weights (4.1B parameters) plus a working margin. A long context needs more.
What is the cheapest GPU to run pi05_telearms_v4_fullft_lr7.07e-5 on?
At 16-bit, 1x MI300X from $1.85 an hour; at 4-bit, 1x MI300X from $1.85 an hour, at the lowest on-demand prices the SAVRN Index lists.
Similar Models
Final checkpoint after 60,000 optimization steps. This is a trained policy, not an evaluation result. Policy weights, policy configuration, preprocessing/postprocessing and normalization states are at the repository root. Training resume files are in trainingstate/. Host-specific paths were removed from JSON metadata; supply local dataset/output paths when resuming. The tokenizer reference points to google/paligemma-3b-pt-224 (training revision 35e4f46485b4d07967e7e9935bc3786aad50687c). For subtask models, supply the corresponding per-frame subtask as the policy task text. For 3-view models also supply the dataset-defined keyframe image. No real-robot evaluation metrics are claimed here.…
Final checkpoint after 60,000 optimization steps. This is a trained policy, not an evaluation result. Policy weights, policy configuration, preprocessing/postprocessing and normalization states are at the repository root. Training resume files are in trainingstate/. Host-specific paths were removed from JSON metadata; supply local dataset/output paths when resuming. The tokenizer reference points to google/paligemma-3b-pt-224 (training revision 35e4f46485b4d07967e7e9935bc3786aad50687c). For subtask models, supply the corresponding per-frame subtask as the policy task text. For 3-view models also supply the dataset-defined keyframe image. No real-robot evaluation metrics are claimed here.…
π₀.₅ (Pi05) is a Vision-Language-Action model from Physical Intelligence designed for open-world generalization: it evolves π₀ to generalize to entirely new environments and situations that were never seen during training. The LeRobot implementation is adapted from their open-source OpenPI repository. This policy has been trained and pushed to the Hub using LeRobot. Learn how to train and run it in the LeRobot pi05 guide, or browse the full documentation. The policy consumes these observation features and produces these action features. Inputs Outputs New to LeRobot? These guides cover the full workflow: - Install LeRobot — set up the lerobot package. - Hardware setup — assemble, wire, and…
π₀.₅ (Pi05) is a Vision-Language-Action model from Physical Intelligence designed for open-world generalization: it evolves π₀ to generalize to entirely new environments and situations that were never seen during training. The LeRobot implementation is adapted from their open-source OpenPI repository. This policy has been trained and pushed to the Hub using LeRobot. Learn how to train and run it in the LeRobot pi05 guide, or browse the full documentation. The policy consumes these observation features and produces these action features. Inputs Outputs New to LeRobot? These guides cover the full workflow: - Install LeRobot — set up the lerobot package. - Hardware setup — assemble, wire, and…
π₀.₅ (Pi05) is a Vision-Language-Action model from Physical Intelligence designed for open-world generalization: it evolves π₀ to generalize to entirely new environments and situations that were never seen during training. The LeRobot implementation is adapted from their open-source OpenPI repository. This policy has been trained and pushed to the Hub using LeRobot. Learn how to train and run it in the LeRobot pi05 guide, or browse the full documentation. The policy consumes these observation features and produces these action features. Inputs Outputs New to LeRobot? These guides cover the full workflow: - Install LeRobot — set up the lerobot package. - Hardware setup — assemble, wire, and…