Together AI
CuratedActiveOpen-model inference API plus self-serve and reserved NVIDIA GPU clusters
From
—
GPU from
$3.99/GPU-hr
Locations
2
countries
Data centers
5
sites
Networks (ASN)
2
ASNs
Plans & prices
4
plans
Overview
Together AI sells serverless and dedicated inference for open-weight models, fine-tuning, and GPU clusters (H100, H200, B200, B300 and GB200/GB300 NVL72) that can be rented hourly or reserved. The San Francisco company was founded in 2022 and builds on its own systems research such as FlashAttention.
Plans & prices
GPU instances4
| Instance | Config | per GPU-hr | Total /hr | Spot | Locations | Source |
|---|---|---|---|---|---|---|
NVIDIA HGX H100 — GPU Clusters on-demand, per GPU | 1× H100 SXM | $3.99 | $3.99 | — | Verified 4 days ago | |
NVIDIA HGX H200 — GPU Clusters on-demand, per GPU | 1× H200 | $5.99 | $5.99 | — | Verified 4 days ago | |
NVIDIA HGX B200 — GPU Clusters on-demand, per GPU | 1× B200 | $8.19 | $8.19 | — | Verified 4 days ago | |
NVIDIA HGX B300 — GPU Clusters on-demand, per GPU | 1× B300 (Blackwell Ultra) | $9.99 | $9.99 | — | Verified 4 days ago |
GPU models
Data centers · 5
Networks (ASN)
Similar providers
falGenerative media platform with pay-per-use model APIs and serverless GPUs
CirrascaleSan Diego AI cloud with NVIDIA, AMD, Qualcomm and Tenstorrent systems on term pricing
Hugging FaceAI model hub selling GPU Spaces, dedicated Inference Endpoints and routed inference
DeepInfraLow-cost model inference API plus dedicated A100-to-B300 GPUs billed by the minute
Fireworks AIFast inference and fine-tuning platform with per-second on-demand GPU deployments
BasetenInference platform for dedicated model deployments and model APIs, billed per minute
GMI CloudNVIDIA Cloud Partner renting H100/H200/B200 and GB200 from its own data centers
GroqInference neocloud built on its own LPU accelerators, serving models via GroqCloud