Unity AI Models and Partners
Unity AI is powered by a variety of artificial intelligence (AI) models that are integrated into the Unity Editor to provide a simple interface and economy for users. Unity AI may change these providers and unique models over time to power, or add to these functionalities. Some of these models are hosted on Unity’s first-party server infrastructure, and some are hosted on third-party infrastructure and accessed via a partner API. This page identifies the AI models used by Unity and third party providers and any additional third party terms of service that may apply.
Assistant
The Unity AI Assistant uses large language models (LLMs) to answer user questions, generate code, and run agentic actions.
| Model and Provider | Purpose | Additional Third Party Terms |
|---|---|---|
GPT-series from Azure Open AI Services | GPT-models are used at various stages in the Assistant pipeline. | n/a |
Llama-series from Meta Platforms, Inc. | Meta Llama-models are used at various stages in the Assistant pipeline. |
Model and Provider
| Model and Provider | Purpose | Additional Third Party Terms |
|---|---|---|
GPT-series from Azure Open AI Services | ||
| Purpose GPT-models are used at various stages in the Assistant pipeline. | ||
| Additional Third Party Terms n/a |
| Model and Provider | Purpose | Additional Third Party Terms |
|---|---|---|
Llama-series from Meta Platforms, Inc. | ||
| Purpose Meta Llama-models are used at various stages in the Assistant pipeline. | ||
| Additional Third Party Terms |
Generators
The Unity AI Generators use several first-party (Unity) and third-party AI models to generate and refine assets. You can read more here about how Unity has trained its first-party models.
| Model Provider | Purpose | Additional Third Party Terms |
|---|---|---|
Sprite | ||
Scenario models include a variety of pre-trained LoRAs or “styles” trained on Stable Diffusion and FLUX foundation models. The models can generate images (“sprites”) from text and can optionally use several image reference types including hand drawn doodles to guide generations. Scenario models also power some post-process refinements. | ||
Layer models include a variety of pre-trained LoRAs or “styles” trained on Stable Diffusion and FLUX foundation models. The models can generate images (“sprites”) from text and several image reference types including hand drawn doodles. Layer models also power (i) post-process refinements, and (ii) custom model training with your choice of foundation models like Stable Diffusion, FLUX, or Bria. *Bria is notably trained on licensed data sources that may offer you further legal protections. | ||
Texture | ||
Unity | Unity first-party model that generates textures from text, image, and pattern references. | n/a |
Animation | ||
Unity | Unity first-party model that generates humanoid animations. This is accessible via the “text-to-animation” prompt. | n/a |
Kinetix model generates humanoid animations from video prompts. This is accessible via the “video-to-animation” prompt. | ||
Sound | ||
Unity | Unity first-party model that generates sound from text and sound prompts. The sound refinements are local operations which are not powered by AI. | n/a |
Model Provider
| Model Provider | Purpose | Additional Third Party Terms |
|---|---|---|
Sprite | ||
| Purpose | ||
| Additional Third Party Terms |
| Model Provider | Purpose | Additional Third Party Terms |
|---|---|---|
| Purpose Scenario models include a variety of pre-trained LoRAs or “styles” trained on Stable Diffusion and FLUX foundation models. The models can generate images (“sprites”) from text and can optionally use several image reference types including hand drawn doodles to guide generations. Scenario models also power some post-process refinements. | ||
| Additional Third Party Terms |
| Model Provider | Purpose | Additional Third Party Terms |
|---|---|---|
| Purpose Layer models include a variety of pre-trained LoRAs or “styles” trained on Stable Diffusion and FLUX foundation models. The models can generate images (“sprites”) from text and several image reference types including hand drawn doodles. Layer models also power (i) post-process refinements, and (ii) custom model training with your choice of foundation models like Stable Diffusion, FLUX, or Bria. *Bria is notably trained on licensed data sources that may offer you further legal protections. | ||
| Additional Third Party Terms |
| Model Provider | Purpose | Additional Third Party Terms |
|---|---|---|
Texture | ||
| Purpose | ||
| Additional Third Party Terms |
| Model Provider | Purpose | Additional Third Party Terms |
|---|---|---|
Unity | ||
| Purpose Unity first-party model that generates textures from text, image, and pattern references. | ||
| Additional Third Party Terms n/a |
| Model Provider | Purpose | Additional Third Party Terms |
|---|---|---|
Animation | ||
| Purpose | ||
| Additional Third Party Terms |
| Model Provider | Purpose | Additional Third Party Terms |
|---|---|---|
Unity | ||
| Purpose Unity first-party model that generates humanoid animations. This is accessible via the “text-to-animation” prompt. | ||
| Additional Third Party Terms n/a |
| Model Provider | Purpose | Additional Third Party Terms |
|---|---|---|
| Purpose Kinetix model generates humanoid animations from video prompts. This is accessible via the “video-to-animation” prompt. | ||
| Additional Third Party Terms |
| Model Provider | Purpose | Additional Third Party Terms |
|---|---|---|
Sound | ||
| Purpose | ||
| Additional Third Party Terms |
| Model Provider | Purpose | Additional Third Party Terms |
|---|---|---|
Unity | ||
| Purpose Unity first-party model that generates sound from text and sound prompts. The sound refinements are local operations which are not powered by AI. | ||
| Additional Third Party Terms n/a |
Inference Engine
The Unity Inference Engine enables you to run AI models on your local machine in the Unity Editor, or on the end-user device in the Unity runtime. No data from these models are stored or transferred to the cloud. Inference Engine does not include built-in models, but rather allows you to import your own custom pre-trained models, or ones acquired from model gardens like Hugging Face. See our documentation for further information and supported workflows.
If you have questions about how Unity AI uses various AI models, please contact legal@unity3d.com.