141GB HBM3e doubles the memory of the H100 while maintaining NVLink 900 GB/s. Fits 70B+ models without sharding. Best for memory-bound workloads like long-context inference and large batch training.

When is the H200 not a good fit?

Limited availability and premium pricing. If your model fits in 80GB or the workload isn't memory-bottlenecked, the H100 offers better price-per-GPU-hour.

Are H200 prices going up or down?

On-demand pricing has increased by about 3% since June 2025, from $3.82 to $3.94/hr per GPU.

What size AI models can the H200 run?

With 141GB of VRAM, the H200 can usually run 70B-class models with headroom, and may handle much larger models in 4-bit quantized form depending on runtime overhead, KV cache, and context length.

How much VRAM does the H200 have?

The H200 has 141GB of VRAM. Multi-GPU setups increase total memory, but that memory is not automatically pooled across GPUs.

What is the H200's memory bandwidth?

The H200 has 4,800 GB/s of memory bandwidth. Higher bandwidth helps with faster data transfer between GPU memory and compute cores.

What data types does the H200 support?

The H200 supports 7 precision formats. Training: BF16, FP16, TF32, FP32. Inference: FP8, INT8. Scientific: FP64.

Does the H200 support NVLink?

Yes. The H200 supports NVLink with 900 GB/s of bidirectional bandwidth. This helps accelerate multi-GPU communication.

How much does the H200 cost per hour?

H200 pricing currently ranges from $1.00/hr to $13.78/hr per GPU, depending on the provider, instance type, and billing model.

How much does the H200 cost per month?

At 720 hours per month, one H200 can cost between $722.41 to $9,921.60 per month, depending on the provider. Reserved and spot pricing can lower that further.

Which cloud providers offer the H200?

The H200 is available from 33 cloud providers, including Runpod, Civo, Cirrascale. Pricing and availability vary by region and billing model.

Can I rent the H200 in the cloud?

Yes. We currently track 168 H200 listings across 33 cloud providers: Billing type Listings Avg $/GPU/hr On-demand 75 $4.00/hr Reserved 56 $3.26/hr Spot 34 $2.96/hr Custom contract 3 $2.10/hr

Nvidia H200

Extends the H100 with doubled memory for large-model inference and training.

Compare vs other GPUs →

Aggregating historical prices...

Key Specifications

Architecture

Hopper

Memory

141GB HBM3e

Memory Bandwidth

4,800 GB/s

Release date

Q4 2023

Compare Cloud Provider Prices

There is a 92% difference between the highest and lowest listings for the H200. You can pay up to $13.78/hr, but the market floor is currently $1.00/hr per GPU (spot instance).

Billing

Max price/hr

Min GPUs

Min vCPUs

Min RAM (GB)

of shown

Default ranking

Our algorithm weighs five factors to find the most relevant matches for you:

Price: We blend total hourly price with price-per-GPU to balance affordability and value.
Specs: We favor offers with higher CPU, RAM, and GPU Memory.
Billing: We favor on-demand billing for simplicity and flexibility over spot instances, reservations, and custom quotes.
Location: We blend both datacenter proximity and provider HQ location. Datacenter location matters for latency, while HQ location matters for compliance and support.
Provider diversity: Each time a provider appears in the list, their subsequent offerings are ranked a little lower, so one provider's offerings don't crowd out the top positions.

Sorting and filtering

Click any column header to sort by that column. Use the filters above the table to narrow results by billing type, GPU count, vCPUs, or RAM. Custom sorting resets the default relevance ranking.

Transparency and funding

Ads and sponsors: Any paid placements are fixed at the top and clearly labeled as sponsored content.

Affiliates: Any affiliate links will be indicated to you as well. We may earn a commission if you click them, but this never influences the ranking order.

Provider	GPUs	Total VRAM	vCPUs	RAM	Billing	$/GPU/h	Total/h	Availability
Vast.ai Our sponsor	1x H200 1x H200 141GB (27574759)	141GB	192	257GB	Spot Variable spot pricing with potential interruptions.	$1.01	$1.01	Available	View
Theta EdgeCloud	1x H200 1x H200 141GB (vm_gh200x1)	141GB	8	141GB	On-Demand Pay-as-you-go pricing. No term commitments.	$2.49	$2.49	Available	View
DigitalOcean	1x H200 SXM 1x H200 141GB SXM (gpu-h200x1-141gb)	141GB	24	240GB	On-Demand Pay-as-you-go pricing. No term commitments.	$3.44	$3.44	Unknown	View
Runpod	1x H200 SXM 1x H200 141GB SXM (community-cloud)	141GB	12	188GB	On-Demand Pay-as-you-go pricing. No term commitments.	$3.59	$3.59	Available	View
Together	1x H200 1x H200 141GB	141GB	--	--	On-Demand Pay-as-you-go pricing. No term commitments.	$4.99	$4.99	Unknown	View
Koyeb	1x H200 1x H200 141GB	141GB	15	240GB	On-Demand Pay-as-you-go pricing. No term commitments.	$3.00	$3.00	Unknown	View
Runpod	1x H200 SXM 1x H200 141GB SXM (secure-cloud)	141GB	12	188GB	On-Demand Pay-as-you-go pricing. No term commitments.	$4.39	$4.39	Available	View
Theta EdgeCloud	2x H200 2x H200 141GB (vm_gh200x2)	282GB	16	282GB	On-Demand Pay-as-you-go pricing. No term commitments.	$2.49	$4.98	Available	View
Civo	1x H200 SXM 1x H200 141GB SXM	141GB	23	169GB	On-Demand Pay-as-you-go pricing. No term commitments.	$3.49	$3.49	Unknown	View
Hyperstack	1x H200 SXM 1x H200 141GB SXM	141GB	22	225GB	On-Demand Pay-as-you-go pricing. No term commitments.	$3.50	$3.50	Unknown	View
Nebius	1x H200 SXM5 1x H200 141GB SXM5	141GB	16	200GB	On-Demand Pay-as-you-go pricing. No term commitments.	$4.50	$4.50	Available	View
Sesterce	1x H200 1x H200 141GB	141GB	44	185GB	On-Demand Pay-as-you-go pricing. No term commitments.	$4.85	$4.85	Available	View
Crusoe	8x H200 8x H200 141GB (h200-141gb-sxm-ib.8x)	1.1TB	176	2TB	On-Demand Pay-as-you-go pricing. No term commitments.	$4.29	$34.32	Unknown	View
FluidStack	1x H200 1x H200 141GB	141GB	48	256GB	On-Demand Pay-as-you-go pricing. No term commitments.	$2.30	$2.30	Unknown	View
Oblivus	1x H200 1x H200 141GB	141GB	224	1.5TB	On-Demand Pay-as-you-go pricing. No term commitments.	$3.99	$3.99	Unknown	View
DigitalOcean	8x H200 SXM 8x H200 141GB SXM	1.1TB	--	--	On-Demand Pay-as-you-go pricing. No term commitments.	$3.44	$27.52	Unknown	View
Runpod	2x H200 SXM 2x H200 141GB SXM (community-cloud)	282GB	24	376GB	On-Demand Pay-as-you-go pricing. No term commitments.	$3.59	$7.18	Available	View
Verda	1x H200 SXM5 1x H200 141GB SXM5 (1H200.141S.44V)	141GB	44	182GB	On-Demand Pay-as-you-go pricing. No term commitments.	$4.00	$4.00	Unknown	View
Koyeb	2x H200 2x H200 141GB	282GB	30	480GB	On-Demand Pay-as-you-go pricing. No term commitments.	$3.00	$6.00	Unknown	View
Theta EdgeCloud	4x H200 4x H200 141GB (vm_gh200x4)	564GB	32	564GB	On-Demand Pay-as-you-go pricing. No term commitments.	$2.49	$9.96	Available	View
No offerings matching your filters.

Heads up: We do our best to keep these specs & prices accurate. However, cloud costs may fluctuate based on region, usage, and other factors not listed here. These are estimates based on common setups and are for informational purposes only. Always verify current rates & exact specs with the provider before provisioning.

Frequently Asked Questions

Why choose the H200?: 141GB HBM3e doubles the memory of the H100 while maintaining NVLink 900 GB/s. Fits 70B+ models without sharding. Best for memory-bound workloads like long-context inference and large batch training.
When is the H200 not a good fit?: Limited availability and premium pricing. If your model fits in 80GB or the workload isn't memory-bottlenecked, the H100 offers better price-per-GPU-hour.
Are H200 prices going up or down?: On-demand pricing has increased by about 3% since June 2025, from $3.82 to $3.94/hr per GPU.
What size AI models can the H200 run?: With 141GB of VRAM, the H200 can usually run 70B-class models with headroom, and may handle much larger models in 4-bit quantized form depending on runtime overhead, KV cache, and context length.
How much VRAM does the H200 have?: The H200 has 141GB of VRAM. Multi-GPU setups increase total memory, but that memory is not automatically pooled across GPUs.
What is the H200's memory bandwidth?: The H200 has 4,800 GB/s of memory bandwidth. Higher bandwidth helps with faster data transfer between GPU memory and compute cores.
What data types does the H200 support?: The H200 supports 7 precision formats. Training: BF16, FP16, TF32, FP32. Inference: FP8, INT8. Scientific: FP64.
Does the H200 support NVLink?: Yes. The H200 supports NVLink with 900 GB/s of bidirectional bandwidth. This helps accelerate multi-GPU communication.
How much does the H200 cost per hour?: H200 pricing currently ranges from $1.00/hr to $13.78/hr per GPU, depending on the provider, instance type, and billing model.
How much does the H200 cost per month?: At 720 hours per month, one H200 can cost between $722.41 to $9,921.60 per month, depending on the provider. Reserved and spot pricing can lower that further.
Which cloud providers offer the H200?: The H200 is available from 33 cloud providers, including Runpod, Civo, Cirrascale. Pricing and availability vary by region and billing model.
Can I rent the H200 in the cloud?: Yes. We currently track 168 H200 listings across 33 cloud providers:

Billing type Listings Avg $/GPU/hr

On-demand 75 $4.00/hr

Reserved 56 $3.26/hr

Spot 34 $2.96/hr

Custom contract 3 $2.10/hr

Billing type	Listings	Avg $/GPU/hr
On-demand	75	$4.00/hr
Reserved	56	$3.26/hr
Spot	34	$2.96/hr
Custom contract	3	$2.10/hr

Technical Specifications

Feature	H200 SXM¹	H200 NVL¹
FP64	34 TFLOPS	34 TFLOPS
FP64 Tensor Core	67 TFLOPS	67 TFLOPS
FP32	67 TFLOPS	67 TFLOPS
TF32 Tensor Core²	989 TFLOPS	989 TFLOPS
BFLOAT16 Tensor Core²	1,979 TFLOPS	1,979 TFLOPS
FP16 Tensor Core²	1,979 TFLOPS	1,979 TFLOPS
FP8 Tensor Core²	3,958 TFLOPS	3,958 TFLOPS
INT8 Tensor Core²	3,958 TFLOPS	3,958 TFLOPS
GPU Memory	141GB	141GB
GPU Memory Bandwidth	4.8TB/s	4.8TB/s
Decoders	7 NVDEC, 7 JPEG	7 NVDEC, 7 JPEG
Confidential Computing	Supported	Supported
Max Thermal Design Power (TDP)	Up to 700W	Up to 600W
Multi-Instance GPUs	Up to 7 MIGs @18GB each	Up to 7 MIGs @18GB each
Form Factor	SXM	PCIe
Interconnect	NVIDIA NVLink™: 900GB/s, PCIe Gen5: 128GB/s	2- or 4-way NVIDIA NVLink bridge: 900GB/s, PCIe Gen5: 128GB/s