Skip to content
SocialAtoZ

Together AI

Verified

AI Acceleration Cloud for inference, fine-tuning, and GPU clusters

AI Development Platform
–

Not yet rated. Be the first to review Together AI.

Visit website Request pricing Write a review
Together AI screenshot See all screenshots
  • Deployment Cloud Based
  • Starting price From $0.03/1M tokens
  • Free trial Not offered
  • Best for Startups, SMEs, Enterprises

What is Together AI?

Together AI is a cloud platform for building and running generative AI. It gives developers API access to more than 200 open source models for chat, vision, image, video, and audio, plus dedicated GPU endpoints, rented GPU clusters, and fine-tuning tools for models such as Llama, DeepSeek, and Qwen. The API is OpenAI-compatible, so existing code often needs only a base URL change to switch providers.

Beyond serverless inference, Together AI offers provisioned throughput, a batch API for lower-cost asynchronous jobs, code sandboxes, and instant or reserved NVIDIA GPU clusters (H100, H200, B200) for training. The company is based in San Francisco and serves developers, startups, and enterprises building production AI applications.

Key Features of Together AI

Together AI lists 9 documented features, including Serverless inference API for 200+ open source models, Dedicated single-tenant GPU endpoints and On-demand and reserved GPU clusters (H100, H200, B200). The list below covers what the product does rather than how it is marketed.

  • Serverless inference API for 200+ open source models
  • Dedicated single-tenant GPU endpoints
  • On-demand and reserved GPU clusters (H100, H200, B200)
  • Fine-tuning for Llama, Mistral, Qwen, and DeepSeek models
  • OpenAI-compatible API for easy migration
  • Batch API at up to 50% lower cost
  • Code sandbox and code interpreter
  • Speculative decoding and FP8 kernel optimizations
  • Voice platform with sub-500ms end-to-end latency

Together AI Pricing

Together AI lists 4 plans without published figures. Pricing is quoted on request.

Serverless Inference

From $0.03/1M tokens

Pay as you go, $5 minimum credit

Request pricing

Per-token pricing across chat, vision, image, video, and audio models, varies by model

Dedicated Endpoints

From $5.49/GPU-hour

Single-tenant GPU

Request pricing

NVIDIA H100 on-demand $5.49/hr, B200 on-demand $8.99/hr

GPU Clusters

From $3.99/GPU-hour

On-demand or reserved

Request pricing

H100 on-demand $3.99/hr; reserved rates as low as $3.09/hr with longer commitments

Enterprise

Custom

Custom

Request pricing

Custom contracts for large scale training and deployment needs

Together AI Specifications

Together AI is available on web app. It offers an API.

Deployment
  • Cloud Based
Billing cycle
Monthly
Desktop
  • Web App
Languages
  • English
Built for
  • Startups
  • SMEs
  • Enterprises
Support
  • Email
  • Tickets
Integrations
OpenAI-compatible API, LangChain, LlamaIndex, Hugging Face
Public API
Yes
Free trial
No
Free plan
No
Runs in browser
Yes
Customisable
Yes
Website
together.ai

Together AI Videos

Together AI Screenshots

Together AI screenshot 1

Together AI Reviews

No reviews yet

Used Together AI? Share your experience and help other buyers decide.

Write the first review

Together AI FAQs

No. Together AI requires a minimum $5 credit purchase to access the platform and does not offer an ongoing free plan.

Together AI hosts more than 200 open source models, including Llama, DeepSeek, Qwen, GLM, and Mistral, for chat, vision, image, video, and audio tasks.

Yes. Together AI's API is OpenAI-compatible, so most existing OpenAI SDK code works by changing the base URL and API key.

Yes. It supports supervised fine-tuning, DPO, and full fine-tuning on major open source model families, priced per million training tokens.

Yes. Together AI offers on-demand and reserved NVIDIA H100, H200, and B200 GPU clusters billed per GPU-hour.