Together AI
Together AI is a cloud platform for running, fine-tuning and serving open-source AI models, from serverless APIs to dedicated endpoints and GPU clusters.
About this tool
Together AI is a cloud built for open-source AI. It hosts many popular open models for text, code, images, video and voice, which developers call through a serverless API without managing infrastructure. The company also publishes research on inference performance and applies it to its own serving stack.
As usage grows, teams can move from serverless calls to dedicated endpoints, provisioned throughput with committed capacity, or batch inference for very large asynchronous jobs. Together also offers fine-tuning and on-demand GPU clusters for training. Serverless usage is priced per token and model size, while dedicated deployments are billed per GPU hour.
What you can do
Serverless inference
Call a catalogue of open models on demand through an API, paying per token without long-term commitments.
Dedicated and provisioned capacity
Reserve dedicated endpoints or committed throughput when traffic is steady and predictable.
Batch inference
Process very large workloads asynchronously at lower cost, with any serverless model or a private deployment.
Fine-tuning and GPU clusters
Train or fine-tune models on your data and rent GPU clusters, including recent NVIDIA hardware, on demand.
Who is it for?
- A startup serves an open model in its product through a serverless API.
- A data team classifies millions of documents with a batch inference job.
- An AI lab rents a GPU cluster to fine-tune a model on proprietary data.
Before you choose
Displayed prices often reflect the cheapest settings; test your real request sizes to estimate costs accurately.
Moving from serverless to dedicated capacity changes billing to GPU hours, which pays off only at sustained volume.
Frequently asked questions
How is Together AI priced?
Serverless usage is charged by tokens and model size, and dedicated endpoints by GPU hour; batch and provisioned options also exist.
Which models does Together AI host?
It hosts many popular open models for text, code, images, video and voice, plus some partner models.
Can I fine-tune models on Together AI?
Yes, Together offers fine-tuning and GPU clusters for training alongside its inference services.
Check the official website for current features, pricing, and terms.
Similar tools
Fireworks AI
CodingFireworks AI is an inference and training platform for open models, offering fast serverless APIs, fine-tuning and dedicated GPU deployments.
Replicate
CodingReplicate runs open-source and proprietary AI models through a cloud API, and lets developers fine-tune models or deploy their own without managing GPUs.
Modal
CodingModal is a serverless cloud for AI workloads: run inference, batch jobs, fine-tuning, sandboxes and GPU notebooks from Python code, billed per second.
Groq
CodingGroq is an inference cloud built around its own LPU chips, offering fast, OpenAI-compatible API access to open AI models for developers and enterprises.
Hugging Face
CodingHugging Face is the main hub for open AI models, datasets and demo apps, with hosting, inference and collaboration tools for teams and enterprises.
Hedra
VideoHedra builds visual AI models and infrastructure, offering managed inference for 100+ models, fine-tuning and GPU sandboxes alongside its creative app.