Models & platforms
Together AI
An AI cloud for inference, fine-tuning, evaluations, GPU compute, and open-model development.
Together AI operates a full-stack AI cloud for serverless and dedicated inference, batch processing, GPU clusters, fine-tuning, evaluations, sandboxes, and managed storage. Its model library emphasizes open and openly available model families.
Key features
- Serverless, batch, provisioned, and dedicated model inference
- Fine-tuning, evaluations, GPU clusters, and development sandboxes
Pros and tradeoffs
Strengths
- Covers inference, customization, and compute in one platform
Consider before choosing
- Infrastructure choices and pricing require workload-specific planning
What people use Together AI for
- Train, customize, and serve open models for production applications
Frequently asked questions
What models can Together AI serve?
Together AI provides a model library and inference services centered on prominent open and openly available model families.
