Pipelines

What is a Pipeline?

A pipeline is a use-case-specific combination of multiple reusable components such as Models, RAGs, Prompts, Guardrails, and Other Pipelines. It processes inputs through a logically structured sequence of these components to generate or predict an output.

Pipelines in GGX can be categorized into two types:

• Chat-based Pipelines: Maintain history and context across multiple user messages and pipeline responses, enabling context-aware interactions.
• Free-Flow Pipelines: Process one input at a time, generating a single output without retaining context from previous interactions.

A typical pipeline consists of:

• Resources: Like Model, RAGs, Prompts, Guardrails, Other Pipelines, etc.
• Pipeline Initialization Logic: Prepares the pipeline for multiple requests if needed.
• Pipeline Scoring Logic: Code that applies the initialized pipelines to provided inputs and uses the various resources to generate/predict responses.

Managing Pipelines on the Platform

The Pipeline Registry organizes all the registered pipelines into customized groups at a centralized location, allowing easier tracking, monitoring, and new pipeline creation.

Registering a Pipeline:

1. Click on Create button in the Pipeline Registry.
2. Fill in important details like Name, Attributes (Output Type, Alias, Pipeline Type), Properties (Group, Permissible Purpose, Description, Approval Workflow)
3. Define Input Arguments or Configs along with their types and default values.
4. Select registered resources (like Model, Global Functions, Prompts, RAGs, etc.) to use in pipeline definition.
5. Select Input Type (Python-based or Custom registration).
6. Upload custom files/models if required and define pipeline logic by writing code in Initialization Logic and Scoring Logic.
7. Add notes and attach documentation if available in the Additional Information section.
8. Lastly, click Save to complete the registration process.

Note:

• The output type for Chat pipelines is fixed as a dictionary {"output": string, "context": custom type}. The context type can be defined in the Formula Section.
• For chat pipelines, there is a default argument called user_message.
• For chat pipelines, history, context, and user_message are available by default, maintaining historical messages in the standard OpenAI format (Struct[content: String, role: String]), context for information retention across turns, and user_message for the current message.

Registered pipelines can be evaluated in the Pipeline Registry, Human Integrated Testing, or used in downstream pipelines. Live monitoring of pipelines can be done Monitoring Module and Annotation Queues.

Benefits of Pipeline Registration:

• Automated tracking and recording of modifications with efficient version upgrades.
• Automatic detection of Permissible Purpose violations.
• Testing and Comparison with other registered pipelines using custom and standardized validation kits.
• Enhances reusability across downstream applications and enables usage tracking with Lineage Tracking.
• Journey to production becomes more transparent and fully auditable, and production monitoring becomes easier.
• Human Integrated Testing and feedback logging for chat-based pipelines.
• Extract ready-to-productionize executable artifacts.
• Better Collaboration for continuous development and testing.