Can I deploy NVIDIA GH200 or NVIDIA RTX A6000 on Runcrate today?

NVIDIA GH200 is not in stock on Runcrate right now; leave your email on its pricing page to register interest. NVIDIA RTX A6000 is on demand in 6 regions from $0.57/hr, billed by the second on most GPUs.

How much VRAM does each GPU have?

NVIDIA GH200 has 96GB HBM3. NVIDIA RTX A6000 has 48GB GDDR6.

GH200 or RTX A6000: which holds a 70B model in 8-bit weights?

It needs 70 GB of 8-bit weights. On the GH200: 1 GPU, 26 GB spare. On the RTX A6000: 2 GPUs (96 GB). Pick the GH200: fits on one GPU.

GPU comparison

GH200 vs RTX A6000.

The GH200 is ahead on memory (96 GB vs 48 GB) and speed (6.4x the FP16 throughput and up to 5.2x the memory bandwidth). The RTX A6000 makes sense when your model fits in 48 GB and its hourly price is lower.

Launch RTX A6000 Reserve a cluster

Rent either on Runcrate

GH200Not in stock right now
Register interest GH200 pricing
RTX A6000from $0.57/hr
$0.012 per GB/hr$0.37 per 100 TFLOPS/hr
Launch RTX A6000 RTX A6000 pricing

Live hourly price, then per GB of memory and per 100 FP16 TFLOPS.

Which is better for

Which to pick for each job. Memory fit and throughput, from the datasheets.

JobGH200RTX A6000Pick

Hold a 70B model in 8-bit weights
70 GB of 8-bit weights
GH200: 1 GPU, 26 GB spare
RTX A6000: 2 GPUs (96 GB)
GH200. Fits on one GPU.
Serve an 8B model in 16-bit weights
16 GB of 16-bit weights
GH200: 1 GPU, 80 GB spare
RTX A6000: 1 GPU, 32 GB spare
GH200. Same GPU count, up to 5.2x the memory bandwidth.
Fine-tune a 70B model with QLoRA
About 46 GB with 4-bit weights
GH200: 1 GPU, 50 GB spare
RTX A6000: 1 GPU, 2 GB spare
GH200. Same GPU count, 6.4x the FP16 throughput.
Train on a full node
FP16 tensor throughput per node
GH200: 1x, 990 TFLOPS
RTX A6000: 2x, 310 TFLOPS, NVLink
GH200. 6.4x the FP16 throughput per GPU.
Throughput per watt
FP16 TFLOPS per 100 W of TDP
GH200: Superchip power, not a GPU TDP
RTX A6000: 52 TFLOPS per 100 W
Either. Not comparable.

Compare another pair

Hopper

NVIDIA GH200

Not in stock right now

Memory: 96 GB HBM3
FP16 tensor: 990 TFLOPS
Bandwidth: up to 4.0 TB/s

Full pricing details

Ampere

NVIDIA RTX A6000

from$0.57/hr

Available now · 6 regions

Memory: 48 GB GDDR6
FP16 tensor: 154.8 TFLOPS
Bandwidth: 768 GB/s

Launch RTX A6000 Reserve a cluster

Full pricing details

At a glance

Spec by spec. The longer bar is the better number.

Memory

NVIDIA GH20096 GB

NVIDIA RTX A600048 GB

Memory bandwidth

NVIDIA GH200up to 4.0 TB/s

NVIDIA RTX A6000768 GB/s

FP16 tensor

NVIDIA GH200990 TFLOPS

NVIDIA RTX A6000154.8 TFLOPS

FP32

NVIDIA GH20067 TFLOPS

NVIDIA RTX A600038.7 TFLOPS

Full specs

The complete spec sheet.

Spec	NVIDIA GH200	NVIDIA RTX A6000
Architecture	Hopper	Ampere
Memory	96 GB	48 GB
Memory type	HBM3	GDDR6
Memory bandwidth	up to 4.0 TB/s	768 GB/s
FP32	67 TFLOPS	38.7 TFLOPS
FP16 tensor	990 TFLOPS	154.8 TFLOPS
INT8	1979 TOPS	310 TOPS
Power	up to 1000 W per superchip (CPU, GPU and memory)	300 W per GPU (TDP)
Form factor	Superchip	PCIe
NVLink	One GPU	Yes
GPUs per node	1	2
Hourly price (live)	Not in stock right now	from $0.57/hr

Questions

GH200 or RTX A6000. What buyers ask first.

Other comparisons