Two models, one decision
Each page puts two policies against each other on the things that decide a project: what card the run needs, how long the control loop is, and how much data it takes before the arm does the task.
- 6 comparisons
- Numbers from the training backend
- One dataset format for all five
GR00T N1.7 vs Pi0.5: speed against contact precision
GR00T N1.7 or Pi0.5 for an SO-100: 152 ms against 485 ms per action step, batch 32 against batch 1, seed or no seed, and which one your task needs.
GR00T N1.7 vs SmolVLA: an 80 GB card against a 24 GB one
GR00T N1.7 needs an 80 GB card at 4 to 12 USD a run, SmolVLA a 24 GB card at 1 to 3. Which one your dataset size justifies, and when the upgrade pays.
ACT vs SmolVLA: specialist against generalist on the same card
ACT and SmolVLA both train on 24 GB for 1 to 3 USD. One is 80 million parameters from scratch, the other 450 million pretrained with language.
Pi0.5 vs SmolVLA: when flow-matching precision pays for the extra hours
Pi0.5 costs four times more per run and twice the latency of SmolVLA. The failure pattern that justifies the upgrade, and the ones that do not.
ACT vs GR00T N1.7: the two extremes of the catalog
ACT is 80 million parameters from scratch at 20 ms. GR00T N1.7 is 3 billion pretrained at 152 ms on an 80 GB card. One question decides which you need.
GR00T N1.7 vs GR00T N1.5: which version to select
N1.7 supersedes N1.5 on every axis that matters. The only reason to select N1.5, and the defaults trap that makes the gap look larger than it is.
All five side by side
The same numbers the training backend uses. Every model reads LeRobot v2.1 datasets, so switching models does not mean recording again.
| Model | GPU | Inference | Default steps | Typical run |
|---|---|---|---|---|
| GR00T N1.7NVIDIA | A100 80 GB or H100 80 GB | about 152 ms | 20,000 | about 4 to 12 USD |
| GR00T N1.5NVIDIA | A100 80 GB or H100 80 GB | about 165 ms | 2,000 | about 4 to 12 USD |
| Pi0.5Physical Intelligence | A100 80 GB or H100 80 GB | about 485 ms | 30,000 | about 4 to 12 USD |
| SmolVLAHugging Face | RTX 4090 or any card with 24 GB | about 245 ms | 20,000 | about 1 to 3 USD |
| ACTStanford (ALOHA) | RTX 4090 or any card with 24 GB | about 20 ms | 100,000 | about 1 to 3 USD |
Prices are estimates: the pool rents GPUs on a spot market and selects them by VRAM, so the hourly rate moves day to day. Details per model under /policies, the runnable instructions under /train.
The arm decides less than you think
A real SO-100 is online and free to drive in the browser. The model you pick matters more than the hardware under it.