Together

Together is an Artificial Intelligence (AI) company that provides cloud-based infrastructure and open-source models for building and running Generative AI (GenAI) systems.

• Cloud platform for training, fine-tuning, and serving large language models (AI infrastructure)
• Hosted access to open-source and custom foundation models via APIs (AI platform)
• Tools for model customization, evaluation, and deployment in enterprise workflows (MLOps / AI lifecycle)
• Support for on-premises (on-prem), Virtual Private Cloud (VPC), and hybrid deployment patterns (hybrid AI infrastructure)
• Focus on cost-efficient compute, optimized inference, and collaboration with the open-source AI ecosystem (AI infrastructure and tooling)

More About Together

Together operates in the enterprise AI infrastructure and platform category, providing services that enable organizations to train, fine-tune, and deploy large language models and other generative models at scale. Its offerings target technical teams that want the flexibility of open-source models with the operational characteristics of a managed cloud platform. Customers use Together to build applications such as chat assistants, code generation tools, and domain-specific language interfaces while retaining control over models and data.

The Together platform (AI infrastructure) exposes capabilities through APIs and SDKs that handle model inference, fine-tuning workflows, and data handling. The service is architected to run across cloud compute resources with support for GPUs and other accelerators, with a focus on cost-aware scheduling and resource allocation. Together’s infrastructure is designed to support large model weights, distributed inference, and high-throughput serving patterns required for production conversational and generative workloads.

Enterprises use Together’s hosted model endpoints (AI platform) to access open-source and commercially usable models without managing low-level deployment details. The platform supports standard interfaces common in the AI ecosystem, such as HTTP-based inference APIs and token-based usage metering, which simplifies integration into existing applications, back-end services, and data pipelines. Organizations can route traffic to different models, experiment with prompts, and observe performance characteristics as they refine their AI use cases.

Together also focuses on model customization and fine-tuning workflows (MLOps / AI lifecycle), enabling enterprises to adapt base models to their own domains and datasets. This includes mechanisms for supervised fine-tuning, prompt engineering support, and evaluation tooling that measures quality, latency, and cost. These capabilities allow teams to iterate on models while maintaining governance over sensitive data and meeting compliance constraints through options such as deployment in VPC environments.

From a marketplace and taxonomy perspective, Together fits into categories including AI Infrastructure-as-a-Service (IaaS), managed Large Language Model (LLM) hosting, and open-source model operations. Its emphasis on open models and flexible deployment aligns with organizations that seek alternatives to closed, single-vendor AI stacks, while still requiring managed reliability, observability, and performance controls for production environments.