Developers who want open models without running hardware; pay per token at low rates, though no free tier is stated.

Kenning's take

What it does

Together AI is a cloud service where developers call open-source models such as DeepSeek, Qwen, GLM, Gemma and Llama through a pay-per-token API. It also rents GPU clusters, offers dedicated endpoints for steady workloads, and provides fine-tuning and custom training. Image, video, speech and transcription models are available on the same platform.

• Serverless pay-per-token inference
• Dedicated endpoints and provisioned throughput
• Fine-tuning and custom post-training
• GPU clusters including B200 and H100
• Image, video, audio and transcription models

Who it suits. Developers and AI companies that want to run open models without managing their own hardware.

Key facts

Category

Developer tools

Made by

Together AI

How to get it

self-serve API sign-up; dedicated capacity and clusters through sales

Works on

API, Web

API price

varies by model; e.g. gpt-oss-120B $0.15 per million input tokens, $0.60 per million output; Llama 3.3 70B $1.04 in and out

FAQ

Facts from the maker's own pages and the sources listed. Checked 2026-10-11.