together-dedicated-containers

What problem does it solve?

Deploy custom Dockerized inference workloads on Together AI's managed GPU infrastructure, enabling teams to run specialized models and pipelines without managing hardware, orchestration, or scaling.

Core Features & Use Cases

• Jig CLI for building, pushing, and deploying: Streamlines container lifecycle from local build to production deployment.
• Sprocket SDK for request handling inside containers: Provides a consistent in-container inference workflow.
• Queue API for asynchronous job submission: Supports prioritized, scalable inference tasks with progress tracking.
• Container Registry access: Private registry hosting images with versioning and security controls.
• Use Case: Deploy custom model servers, multimedia generation pipelines (image/video), or other GPU-accelerated workloads beyond standard endpoints.

Quick Start

Build and deploy a minimal container workflow using Jig CLI to verify end-to-end setup.