H200 pricing on Together AI
Every Together AI H200 offering we track, normalized to a per-GPU hourly rate and ranked against 28 providers with current H200 pricing.
The cheapest H200 we currently track is $1.40/hr on Verda.
Together AI H200 rates
Hourly prices per GPU. Multi-GPU rows show the per-GPU rate, not the instance total.
| Price / GPU-hr | Type | GPUs |
|---|---|---|
| $6.60 | On-demand | 1 |
| $6.60 | On-demand | 2 |
| $6.60 | On-demand | 4 |
| $6.60 | On-demand | 8 |
H200 specifications
- VRAM
- 141GB
- Architecture
- Hopper
- CUDA Cores
- 16,896
- Tensor Cores
- 528
- Memory Bandwidth
- 4800 GB/s
- TDP
- 700W
H200 on other providers
- Verda$1.40/hr
- fal.ai$2.10/hr
- Nebius$2.45/hr
- GMI Cloud$2.60/hr
- CoreWeave$2.62/hr
- Deep Infra$2.69/hr
- Civo$2.99/hr
- Seeweb$3.00/hr
- Koyeb$3.00/hr
- Massed Compute$3.44/hr
- Atlas Cloud$3.50/hr
- Runpod$3.59/hr
- Theta EdgeCloud$3.69/hr
- IO.NET$3.79/hr
- Denvr Dataworks$3.87/hr
- Hyperbolic$3.99/hr
- Hyperstack$3.99/hr
- Jarvis Labs$3.99/hr
- Oblivus$3.99/hr
- AceCloud$4.02/hr
- Crusoe$4.29/hr
- DigitalOcean$4.47/hr
- Modal$4.54/hr
- Sesterce$4.85/hr
- Amazon AWS$7.91/hr
- Oracle Cloud$10.00/hr
- Microsoft Azure$10.60/hr
About Together AI
Together AI is the AI Native Cloud platform engineered for developers building with open-source and frontier AI models. They provide serverless inference, fine-tuning, and GPU clusters with industry-leading performance optimizations.
Frequently Asked Questions
How much does the H200 cost on Together AI?
Together AI lists the H200 from $6.60 per GPU-hour across 4 tracked configurations. Prices are collected daily — see the table above for the current rates.
Is Together AI the cheapest place to rent the H200?
Together AI ranks #25 of 28 providers we track with current H200 pricing. The cheapest right now is Verda. Price is only one factor — region availability, network, and storage costs differ between providers.
Which H200 configurations does Together AI offer?
We currently track 4 H200 offerings from Together AI, covering the GPU counts, pricing types, and regions listed in the table above. Every price is normalized to a per-GPU hourly rate so it can be compared across providers.