DEDUPEDWhat do you want to do?
3 use cases
-
Add a test-time training memory to a model
Wrap attended action chunks with TTTWrapper and an MLP memory in a few lines of PyTorch.
View prompt
Install lucidrains/robo_ttt, run the basic MemoryKeyValueBind plus TTTWrapper example on random tensors of shape (2, 5, 4, 512), and confirm the output shape matches the input.
-
Wrap a robot policy model with RoboTTT
Use the full RoboTTT wrapper around a policy such as MimicVideo.
View prompt
Wrap a MimicVideo policy with RoboTTT from lucidrains/robo_ttt, run one forward and backward pass on dummy data, and confirm gradients flow.
-
Supervise actions with next-latent prediction
Enable predict_next_latent so action generation is supervised by next-state representation prediction.
View prompt
In robo_ttt, enable predict_next_latent with a state_arg, train for a few steps on dummy data, and report the dynamics loss. Confirm rollout returns the predicted next latent.
Quick start with Robo TTT
- Check you have what it needs: Python with PyTorch.
- Open lucidrains/robo_ttt on GitHub and read the README and license.
- Install with
pip install robo-ttt - Build a 2-layer MLP memory and wrap it in MemoryKeyValueBind
- Create TTTWrapper(512, memory = memory)
- Pass action chunks and check the output shape
These steps come from our top use case, Add a test-time training memory to a model, which also has a copy-paste agent prompt.
Use this in your AI agent
Ask your agent
Using the Deduped MCP, get the setup steps for lucidrains/robo_ttt
Add Deduped to Claude Code once
claude mcp add --transport http deduped https://dedupedai.com/mcp
Every Monday: the 5 hottest repos and exactly how to use them. Free, one email a week, unsubscribe anytime.
Get them freeRobo TTT alternatives · Best AI repos for Security and science
About this repo
- Who it is forRobotics and ML researchers experimenting with test-time training in robot policies.
- Why it is hot105 stars, launched 90 days ago
- LanguagePython
- Sourcegithub.com/lucidrains/robo_ttt · Back to Hot repos
Requirements
- Python with PyTorch
Add this badge to your README
Maintain Robo TTT? Show readers where to find these use cases. 
Markdown
[](https://dedupedai.com/use-cases/lucidrains-robo-ttt/?utm_source=github&utm_medium=badge&utm_campaign=lucidrains-robo_ttt)
HTML
<a href="https://dedupedai.com/use-cases/lucidrains-robo-ttt/?utm_source=github&utm_medium=badge&utm_campaign=lucidrains-robo_ttt"><img src="https://dedupedai.com/badge/lucidrains/robo_ttt.svg" alt="Use cases on Deduped" height="20"></a>
A static image, no tracking script. About the badge
Questions about Robo TTT
What is Robo TTT?
robo_ttt is a PyTorch implementation of RoboTTT (Jiang et al., Stanford and Nvidia), providing test-time training wrappers and memory modules for robot policy models, including optional next-latent forward dynamics. It is an open-source project on GitHub: https://github.com/lucidrains/robo_ttt.
Who is Robo TTT for?
Robotics and ML researchers experimenting with test-time training in robot policies.
What can I do with Robo TTT?
3 use cases so far, for example: Add a test-time training memory to a model; Wrap a robot policy model with RoboTTT; Supervise actions with next-latent prediction. Each has steps and a copy-paste agent prompt.
What do I need to run Robo TTT?
Python with PyTorch.
What are alternatives to Robo TTT?
Other hot AI repos in Security and science: Security Audit Skill, Agentbaseline, RiftX, Hacker, Codex Privacy Hud.
Is Robo TTT safe to run?
Deduped does not own, audit or vouch for it. Read the README and license, run it on your own systems first and review every command before you run it.