PaLM-SayCan
No estimate
No hardware requirements for this model
The weights for this model have not been published, so it cannot be downloaded or run on your own hardware at any size. It is reachable only through its provider, and no graphics card changes that.
On record
Full specification
Everything on record for this model. Most of it describes how it was trained rather than how it runs — useful context for judging how much work went into it, and how it compares with models built at a different scale.
Origin
Who built this model, where, and when it was published.
- Organisation
- Organisation type
- Industry
- Country
- United States of America
- Published
- 16 August 2022
- Authors
- Michael Ahn, Anthony Brohan, Noah Brown, Yevgen Chebotar, Omar Cortes, Byron David, Chelsea Finn, Chuyuan Fu, Keerthana Gopalakrishnan, Karol Hausman, Alex Herzog, Daniel Ho, Jasmine Hsu, Julian Ibarz, Brian Ichter, Alex Irpan, Eric Jang, Rosario Jauregui Ruano, Kyle Jeffrey, Sally Jesmonth, Nikhil J Joshi, Ryan Julian, Dmitry Kalashnikov, Yuheng Kuang, Kuang-Huei Lee, Sergey Levine, Yao Lu, Linda…
What it does
The problem areas the model was built for. A model can carry several of each.
- Domain
- Robotics
- Task
- Robotic manipulation
- Approach
- Reinforcement learning
- Base model
- PaLM (540B)
Size
How large the model is and how much data it was trained on. Parameters are the figure that decides whether it fits on a given graphics card.
- Parameters
- 540B
- Training data
- tokens
540B from PaLM-540B.
Training compute
The arithmetic performed to train the model, measured in floating-point operations. It is a measure of what the training run cost, not of how fast the finished model answers you.
- Fine-tuning compute
- 2.8 × 10²⁰ FLOP
I think PaLM-540B wasn't actually involved in training and the authors trained a separate RL model. "The RL model is trained using 16 TPUv3 chips and for about 100 hours, as well as a pool of 3000 CPU workers to collect episodes and another 3000 CPU workers to compute target Q-values." Ignoring CPU calculations: 1600 hours * 3600 * 123 teraflops * 0.4 (assumed utilization) = 2.83e20
The training run
What it physically took to train: which chips, how many, for how long, and what that drew from the wall.
- Training hardware
- Google TPU v3
- Chips used
- 16
- Wall-clock time
- 100 hours
- Power draw
- 14.4 kW
The RL model is trained using 16 TPUv3 chips and for about 100 hours
Availability
Whether you can obtain the model and run it on your own hardware, which is what decides if any of the graphics-card figures on this page apply.
- Weights
- Closed — provider access only
- Model access
- Unreleased
- Training code
- Open source
We open source a version of SayCan that works with a simulated tabletop environment. [tabletop saycan] https://github.com/google-research/google-research/tree/master/saycan License: Apache 2.0
How it is classified
Labels the source dataset applies when tracking notable models, and how confident it is in the entry.
- Foundation model
- Yes
- Likely above 10²³ FLOP
- Yes
- Record confidence
- Likely
- Citations
- 3,081
Sources
Where this record came from and when it was last checked.
- Reference
- Do As I Can, Not As I Say: Grounding Language in Robotic Affordances
- Last updated
- 25 May 2026
What the numbers mean
Background
PaLM-SayCan was published by Google, in the country recorded as United States of America, during August 2022. The category the publisher falls under is industry.
It works in the domain of Robotics, and is recorded as performing the task of robotic manipulation.
Its starting point was an existing base model, PaLM (540B). Most models at this scale are adapted from an existing base rather than built from nothing.
Its weights were never published, so it can only be reached through its provider. No graphics card changes that.
Answers
PaLM-SayCan — common questions
PaLM-SayCan— when was it released?
It was published in August 2022. Capability per parameter has improved considerably since, so a newer model of the same size is often the better use of the same hardware.
PaLM-SayCan— what is it used for?
It works in the domain of Robotics, and is recorded as handling the task of robotic manipulation. Models frequently carry more than one of each, and the tags describe purpose rather than capability limits.
PaLM-SayCan— what GPU do I need to run it?
None. This is a closed model — its weights were never published, so it cannot be downloaded or run on your own hardware at any price. It is reachable only through its provider.
PaLM-SayCan— is it open source?
No. Its weights have not been published, so it exists only as a service controlled by its owner.
PaLM-SayCan— how many parameters does it have?
It has a parameter count of 540B. 540B from PaLM-540B. That figure is the total, and it is what decides how much memory the model needs — roughly half a gigabyte per billion at the compression most people use.
PaLM-SayCan— who created it?
It was published by Google, based in United States of America, an organisation categorised as industry.
The other direction
Looking at it from the other side?
This page starts from the model. If you already own a card and want to know everything it will run, start from the hardware instead.