What is Griptape
Griptape enables studios to orchestrate multiple AI models and agents within secure, professional workflows, connected with Foundry’s existing creative toolsets. It enables controlled access to rapidly evolving AI models, balancing innovation with the strict security and traceability needs of major production environments.
What is AI orchestration
AI orchestration is the method of connecting multiple AI models into repeatable workflows for faster, more consistent results within a single, unified system.
Why Griptape
Built for the way you work
Studio-friendly license with easy-to-customize models. Stay on top of the latest developments in AI without breaking your workflows or pipeline.
Studio-grade security
Deploy Griptape securely in the cloud or via an on-premise admin server, where you can keep all your IP and model execution behind your firewall.
Controllable, trackable, scalable, and flexible
Use proven workflows for your pipeline or customize and build your own with Python. Griptape also provides full versioning and control for reliability and consistency.
Get started quickly
With a familiar node-based UI, you can make the most of AI for your creative processes, with a node library for inspiration, and assistants to automate your technical tasks.
Griptape Enterprise: Built for your unique studio needs
Meet the stringent control and security needs of high-end media productions with Griptape Enterprise.
Griptape Enterprise ensures every workflow remains vetted, compliant, and production-ready. Studio-controlled licensing, and expanded project management capabilities enable you to curate which models artists use.
Studios can also connect to Griptape Enterprise through the cloud or via an on-premise admin server, with Griptape Nodes workflows and model execution staying behind your firewall.
Griptape in action
Explore how we’re innovating with Griptape. Check out the latest workflows and developments in our ever-evolving playlist.
Features
Studio-friendly
Licenses built for the way you operate, with the option to control user permissions.
Secure
Operate Griptape on-premise and behind your firewall, giving full peace of mind and visibility to the work done in your instance.
Traceability
Full visibility and transparency into workflows run with AI, ensuring fully controllable, repeatable processes.
Scalable
Expand users and inputs easily and securely. Griptape is designed to grow with you.
Flexible
Edit prompts from the node library or create your own to build workflows tailored for your unique needs.
Artist-friendly UI
The familiar node-based UI enables artists to get started quickly with Griptape.
Pipelineability
Griptape securely and easily integrates into your existing pipelines, reducing friction and enabling you to get started sooner.
FAQs
What is Griptape?
Griptape is an AI orchestration platform which enables artists and studios to orchestrate multiple AI models and agents within secure, professional workflows, connected with Foundry’s existing creative toolsets. It comprises Griptape Nodes, Griptape Framework and Griptape Engine.
What is Griptape Nodes?
Griptape Nodes is a visual interface that enables artists to build AI-driven workflows without writing code. Griptape Nodes is available as a browser interface or a desktop application for Windows, Mac or Linux.
How do Griptape credits work?
Griptape credits allow you to flexibly manage your consumption of third-party cloud models within your creative workflows.
Third-party cloud-hosted models require Griptape credits to run. The Professional subscription includes an allocation of 1,000 credits per month to get you started. If you exhaust your monthly credit volume, you can easily purchase additional credits within the Griptape application.
Running open-source models locally or using cloud models where you bring your own external API keys will not consume your Griptape credits.
What is the Griptape Framework and Griptape Engine?
The Griptape Framework is a Python toolkit enabling developers to build secure GenAI applications and agents that connect to existing studio pipelines. This includes the Griptape Engine which orchestrates the execution of workflows, and can be run either in the cloud or on-premises.
Does Griptape support on-premises deployment?
Yes. Both Griptape Engine and Griptape Nodes can be run either in the cloud or on-premises, allowing studios to run AI automation locally to meet strict security requirements.
What software and hardware is needed to run Griptape?
Griptape Nodes is available either via a web browser or as a desktop application for Windows, Mac or Linux. There are no specialist hardware requirements for building workflows or using cloud-based AI models, but to run AI models locally, you will require a professional-grade GPU.
Which AI models are supported by Griptape?
Griptape is model-agnostic, supporting both open-source and proprietary models for text, image, video, 3D, and audio generation. Griptape includes standard integrations with many leading model vendors including Google, Black Forest Labs, ByteDance, OpenAI, Alibaba, ElevenLabs, Topaz Labs, Luma, Rodin, and Kuaishou. Customers can also connect with a wide range of models from HuggingFace, or bring their own diffusion models. You can see the models Griptape supports here.
How does Griptape integrate with current VFX & Animation pipelines?
Griptape is a Python-based framework, enabling immediate integration with studio pipelines. Griptape also supports MCP, enabling agentic workflows with common DCCs such as Nuke, Maya, and Blender, and with production management systems like Flow Production Tracking.
What is Foundry’s policy on the copyright status and training data of models available in Griptape?
Our goal with Griptape is to empower our customers to experiment and test best-in-class models that align with their own risk tolerance. To provide this flexibility, Griptape supports a wide range of both open-source and commercial models. Because AI is a fast-moving space with a wide variety of training data methods utilized by different model developers, users must verify the safety, compliance, and licensing of each model for their specific use cases. We are also monitoring ongoing legal and industry developments closely.