Skip to content
RobothecaBeta
Log in

Giga-World-1

  • AI model
  • Humanoids and robot learning

An open-source toolkit for training, running and adapting robot world models, used to evaluate robot policies by generating video rollouts.

Made by GigaAI, China (checked on github.com/open-gigaai (opens in a new tab)).

Source on GitHub: github.com/open-gigaai/giga-world-1 (opens in a new tab). Facts read at commit 0ad5e2b (opens in a new tab).

Licenses

Code
Apache-2.0 (opens in a new tab)

Read by Robotheca from LICENSE (opens in a new tab) (Apache-2.0) at commit 0ad5e2b (opens in a new tab), on 16 Oct 2026.

Commercial use
Yes

What it does

  • Provides Stage-1 controllable pre-training and Stage-2 DMD2 distillation for Nano (1.3B) and Pro (5B) models.
  • Offers one-click inference scripts for image-to-video and text-to-video generation at 10 FPS.
  • Includes a data preprocessing pipeline that converts LeRobot-style data into its training format, with a toy dataset.
  • Ships tools for merging LoRA checkpoints into standalone transformers and for downloading weights from Hugging Face or ModelScope.
  • Stage-2 distilled weights, RL post-training and other-domain weights are listed as coming soon.
  • WMBench benchmark is only partially open-sourced.

Written by Robotheca from the project's README (opens in a new tab) at commit 0ad5e2b.

Getting started

Difficulty
Advanced
Hardware it needs, as the project states it
Production setup is a single node with 8 x H20 or 8 x A100 GPUs. Inference and training can run on consumer-grade GPUs with memory-saving settings. Linux, verified on Ubuntu 20.04 and 22.04, with CUDA 12.x recommended.

Maintenance

Status
Maintained: a commit in the last 12 months
Newest commit
12 Jul 2026 on main
Latest release
None published on GitHub

Checked on GitHub on 16 Oct 2026.

Links

Robots on Robotheca that use it

No robot on Robotheca is linked to it yet. Links come from a robot's robot.yaml, from its dependency files read as text, or from Robotheca's editors.

Robotheca lists AI models as a reference: nothing is hosted or run here, and every fact links to where it was read.