# pi0_base by LeRobot: Open-Weight Model Source: https://savrn.com/models/pi0-base Markdown alternate of the page above; the site index is https://savrn.com/llms.txt --- ## Runs On What it takes to serve pi0_base (3.5B 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 | 7.0 GB | 8.4 GB | 1x [MI300X](https://savrn.com/ai-index/pricing/gpus/mi300x) (192 GB) Vultr | $1.85 | [1x H100](https://savrn.com/ai-index/pricing/gpus/h100) $1.99 · [1x MI325X](https://savrn.com/ai-index/pricing/gpus/mi325x) $2.00 | | 8-bit | 3.5 GB | 4.2 GB | 1x [MI300X](https://savrn.com/ai-index/pricing/gpus/mi300x) (192 GB) Vultr | $1.85 | [1x H100](https://savrn.com/ai-index/pricing/gpus/h100) $1.99 · [1x MI325X](https://savrn.com/ai-index/pricing/gpus/mi325x) $2.00 | | 4-bit | 1.8 GB | 2.1 GB | 1x [MI300X](https://savrn.com/ai-index/pricing/gpus/mi300x) (192 GB) Vultr | $1.85 | [1x H100](https://savrn.com/ai-index/pricing/gpus/h100) $1.99 · [1x MI325X](https://savrn.com/ai-index/pricing/gpus/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](https://savrn.com/ai-index/pricing/gpus), read Oct 22, 2026. [pi0_base on every accelerator the SAVRN Index prices, at every precision](https://savrn.com/models/pi0-base/gpus) ## Model Card π₀ is a Vision-Language-Action (VLA) foundation model from Physical Intelligence that jointly reasons over vision, language, and actions to control robots, serving as the base architecture that later enabled π₀.₅’s open-world generalization. Original paper: π0: A Vision-Language-Action Flow Model for General Robot Controlion For full installation details (including optional video dependencies such as ffmpeg for torchcodec), see the official documentation: https://huggingface.co/docs/lerobot/installation If you’re training / fine-tuning, you typically call forward(...) to get a loss and then: - -policy.chunksize=... - -policy.nactionsteps=... - -policy.maxactiontokens=...… Excerpt from the card by LeRobot, licensed gemma. ## Identity and Version Repository lerobot/pi0_base Publisher LeRobot Task Robotics Modality Control Library lerobot Parameters 3.5B parameters Languages en Revision 25c379b52ba2ff8788cab921758a3cc3fe3f77f2 First published 2025-09-09 Last updated 2026-06-03 ## Files and Weights 6 files, 14.0 GB in total. The weights are 1 file totalling 14.0 GB in safetensors. Weights1 file · 14.0 GB Configuration3 files · 3.7 KB Documentation1 file · 4.7 KB Repository1 file · 1.5 KB Every file | File | Type | Size | SHA-256 | | --- | --- | --- | --- | | model.safetensors | Weights | 14.0 GB | 8229fd9a7c3c | | config.json | Configuration | 1.9 KB | — | | policy_postprocessor.json | Configuration | 565 B | — | | policy_preprocessor.json | Configuration | 1.2 KB | — | | README.md | Documentation | 4.7 KB | — | | .gitattributes | Repository | 1.5 KB | — | ## License and Download License gemma Access Open weights, no gate Download size 14.0 GB [Download from LeRobot](https://huggingface.co/lerobot/pi0_base) Released by LeRobot through its official repository on Hugging Face. ## Evaluations Each result is shown as reported, with the conditions its reporter stated. None is a SAVRN measurement. A comparison lines two results up only when their configuration, unit and setup are all stated and identical. | Benchmark | Conditions | Result | Reported by | Revision | Date | | --- | --- | --- | --- | --- | --- | | VLABench/vlabench_primitive_ft_lerobot_video | Task overallMetric overallSetup Reproduction from ACoT-VLA paper (Table 3). A separate X-VLA paper reproduction reports 37.8.Comparison conditions not established | 44.5 | VLA Model Leaderboard Reported by a third party | Evaluated revision not stated | 2026-03-03 | ## Memory Requirements | Precision | Weights in memory | | --- | --- | | As published | 14.0 GB | | 16-bit | 7.0 GB | | 8-bit | 3.5 GB | | 4-bit | 1.8 GB | Weights only, from the published parameter count; the key-value cache and runtime add to this. ## Built on This Model - Derived from[train_800_sparse__point__overlay_a75__sim__all_cameras__live__pi0__seed_0](https://savrn.com/models/train-800-sparse-point-overlay-a75-sim-all-cameras-live-pi0-seed-0) ## Questions About pi0_base ### How much GPU memory does pi0_base need? About 8.4 GB at 16-bit and 2.1 GB at 4-bit: the weights (3.5B parameters) plus a working margin. A long context needs more. ### What is the cheapest GPU to run pi0_base 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. ### Can I use pi0_base commercially? Yes, with conditions. pi0_base is released under Gemma Terms of Use. Gemma models are released under Google's Gemma Terms of Use, which permit commercial use and redistribution subject to the Gemma Prohibited Use Policy, whose restrictions must be passed on to anyone the model is distributed to. ## Similar Models Model · Robotics ### [pi05_libero_finetuned_v044](https://savrn.com/models/pi05-libero-finetuned-v044) [LeRobot](https://savrn.com/model-publishers/lerobot) π₀.₅ is a Vision-Language-Action (VLA) model with open-world generalization from Physical Intelligence, co-trained on robot demonstrations and large-scale multimodal data to execute long-horizon tasks in unseen real-world environments. Checkpoint trained and evaluated on LIBERO tasks Note: This model currently supports only the flow-matching action head for π₀.₅ training and inference. Other components from the original work (e.g., subtask prediction, action tokenization, or RL) were not released upstream and are not included here, though the LeRobot team is actively working to support them. Original paper: π0.5: A Vision-Language-Action Model with Open-World Generalization For full… Open weights gemma 3.6B parameters lerobot [View model](https://savrn.com/models/pi05-libero-finetuned-v044) Model · Robotics ### [pi05_base](https://savrn.com/models/pi05-base) [LeRobot](https://savrn.com/model-publishers/lerobot) π₀.₅ is a Vision-Language-Action (VLA) model with open-world generalization from Physical Intelligence, co-trained on robot demonstrations and large-scale multimodal data to execute long-horizon tasks in unseen real-world environments. Note: This model currently supports only the flow-matching action head for π₀.₅ training and inference. Other components from the original work (e.g., subtask prediction, action tokenization, or RL) were not released upstream and are not included here, though the LeRobot team is actively working to support them. Original paper: π0.5: A Vision-Language-Action Model with Open-World Generalization For full installation details (including optional video… Open weights gemma 3.6B parameters lerobot [View model](https://savrn.com/models/pi05-base) Model · Robotics ### [pi05_libero_base](https://savrn.com/models/pi05-libero-base) [LeRobot](https://savrn.com/model-publishers/lerobot) π₀.₅ is a Vision-Language-Action (VLA) model with open-world generalization from Physical Intelligence, co-trained on robot demonstrations and large-scale multimodal data to execute long-horizon tasks in unseen real-world environments. Note: This model currently supports only the flow-matching action head for π₀.₅ training and inference. Other components from the original work (e.g., subtask prediction, action tokenization, or RL) were not released upstream and are not included here, though the LeRobot team is actively working to support them. Original paper: π0.5: A Vision-Language-Action Model with Open-World Generalization For full installation details (including optional video… Open weights gemma 3.6B parameters lerobot [View model](https://savrn.com/models/pi05-libero-base) Model · Robotics ### [openpi-pi07-so101-left-arm-multitask](https://savrn.com/models/openpi-pi07-so101-left-arm-multitask) [Pham Tri Quang](https://savrn.com/model-publishers/twanghcmut) Open weights other 3.6B parameters lerobot [View model](https://savrn.com/models/openpi-pi07-so101-left-arm-multitask) Model · Robotics ### [actuator_unboxing_speedcond_t2_fullft_bs256](https://savrn.com/models/actuator-unboxing-speedcond-t2-fullft-bs256) [Dream Machines](https://savrn.com/model-publishers/dreammachines) Full fine-tune of lerobot/pi05base on a bimanual actuator-unboxing task, with a speed token in the text prompt: Part of a check whether pi0.5 can be conditioned through text tokens, as a precursor to advantage conditioning (RECAP, π0.6). Data (not on the Hub): 198 teleoperated episodes labelled slow, merged with 437 episodes of the same task demonstrated about 1.75× faster in another session, labelled fast (635 episodes, 371,523 frames; 50 fps, three 224×224 cameras, 14-D state and action). 30 % of training samples get unknown, the null prompt for classifier-free guidance. Training: 4,400 steps (about 3 epochs), batch 256, learning rate 5e-5, 10 % linear warm-up, cosine decay over the last… Open weights 3.4B parameters lerobot [View model](https://savrn.com/models/actuator-unboxing-speedcond-t2-fullft-bs256) Model · Robotics ### [actuator_unboxing_speedcond_t1_fullft_bs256](https://savrn.com/models/actuator-unboxing-speedcond-t1-fullft-bs256) [Dream Machines](https://savrn.com/model-publishers/dreammachines) Full fine-tune of lerobot/pi05base on a bimanual actuator-unboxing task, with a speed token in the text prompt: Part of a check whether pi0.5 can be conditioned through text tokens, as a precursor to advantage conditioning (RECAP, π0.6). Data (not on the Hub): 198 teleoperated episodes (50 fps, three 224×224 cameras, 14-D state and action) labelled slow, plus two 2× copies of every episode that keep only the even or only the odd frames, labelled fast (594 episodes, 348,616 frames). Slow and fast samples show the same images, so only the token tells them apart. 30 % of training samples get unknown, the null prompt for classifier-free guidance. Training: 4,000 steps (about 3 epochs), batch… Open weights 3.4B parameters lerobot [View model](https://savrn.com/models/actuator-unboxing-speedcond-t1-fullft-bs256) ## LeRobot [All models and datasets](https://savrn.com/model-publishers/lerobot) ## Versions - [25c379b52ba2](https://savrn.com/models/pi0-base/versions/25c379b52ba2) · current 2026-09-18 ## Explore More - [All robotics models](https://savrn.com/models/tasks/robotics) - [All models under gemma](https://savrn.com/models/licenses/gemma) - [Model comparisons](https://savrn.com/models/comparisons) - [The model directory](https://savrn.com/models) - [Open model prices by host](https://savrn.com/ai-index/pricing/open-models) ## Source - Repository metadata, read 2026-10-20. - [Hugging Face record](https://huggingface.co/lerobot/pi0_base) - [How the hub is built](https://savrn.com/model-hub/methodology)