Together AI

Together AI is a cloud platform for building and running AI applications on open source models. It covers inference, GPU compute and model customization in one stack, so teams can move from experimentation to production scale without managing infrastructure. The platform is aimed at developers and AI teams. Serverless inference runs open models on demand across chat, vision, image, audio, video, transcription, embeddings, reranking and moderation. Batch inference processes large asynchronous workloads, provisioned throughput offers reserved capacity with a 99 percent uptime SLA, and dedicated model and container inference provide private deployments for text and generative media. Accelerated compute ranges from self serve instant GPU clusters to thousands of GPUs, and the platform also offers fine tuning, code sandboxes for agents and managed storage with zero egress fees. Serverless inference is priced per million tokens and varies by model. Published examples include MiniMax M3 at $0.30 input and $1.20 output, GLM 5.3 Flash at $0.15 input and $0.50 output, Gemma 4 31B at $0.39 input and $0.97 output, and gpt oss 120B at $0.15 input and $0.60 output. Cached input is billed at a lower rate on supported models, a Batch API price is listed separately, and image models are priced per image, for example $0.053 for GPT Image 2. Prices shown reflect the lowest resolution or duration settings and may vary. Dedicated endpoints, clusters and enterprise terms are arranged with the sales team.