Standard_NV8as_v4 specs and pricing | Azure

Azure Virtual Machine: NV8as_v4 / NV8as v4 with 8 vCPUs and 28 GiB of memory. Available in 25 regions starting from $340.18 per month. A -17.17% cheaper alternative is available.

Name	Standard_NV8as_v4
Details	Standard is recommended tier N – GPU enabled V – Remote visualisation workloads and other graphics-intensive applications 8 – The number of vCPUs a – AMD-based processor s – Premium Storage capable v4 – version
vCPUs	8
CPU Architecture	x64
Memory (GiB)	28
Hyper-V Generations	V1,V2
ACUs	0
GPUs	1
GPU Name	AMD Radeon Instinct MI25
GPU RAM (GiB)	16
GPU Total RAM (GiB)	16
Max Network Interfaces	4
RDMA Enabled	no
Accelerated Net	yes

OS Disk Size	1023 GiB
Res Disk Size	176 GiB
Max Disks	16
Support Premium Disk	yes
Combined IOPS is a sum of all attached disk's IOPs	16000
Uncached Disk IOPS	12800
Combined Write is a sum of all attached disk's write throughtput	128 MiB/Sec
Combined Read is a sum of all attached disk's read throughtput	128 MiB/Sec

Regional Prices

US Dollar ($)

Per Hour

Standard

Pay-as-you-go

Region Name	Region ID	Linux Price	Windows Price
East US	eastus	0.4660	0.8340
East US 2	eastus2	0.4660	0.8340
hidden-1	hidden-1	0.4660	0.8340
West US 2	westus2	0.4660	0.8340
hidden-2	hidden-2	0.5130	0.8810
hidden-3	hidden-3	0.5590	0.9270
hidden-4	hidden-4	0.5590	0.9270
hidden-5	hidden-5	0.5590	0.9270
hidden-6	hidden-6	0.5590	0.9270
hidden-10	hidden-10	0.5820	0.9500
hidden-11	hidden-11	0.5820	0.9500
hidden-12	hidden-12	0.5820	0.9500
hidden-13	hidden-13	0.5820	0.9500
hidden-14	hidden-14	0.5820	0.9500
hidden-7	hidden-7	0.5820	0.9500
hidden-8	hidden-8	0.5820	0.9500
hidden-9	hidden-9	0.5820	0.9500
hidden-15	hidden-15	0.6060	0.9740
hidden-16	hidden-16	0.6060	0.9740
hidden-17	hidden-17	0.6290	0.9970
hidden-18	hidden-18	0.6520	1.0200
hidden-19	hidden-19	0.6520	1.0200
hidden-20	hidden-20	0.6520	1.0200
hidden-21	hidden-21	0.6760	1.0440
hidden-22	hidden-22	0.6760	1.0440

Best AI models you can run on this instance

Top open-weight chat models ranked by Intelligence Index that fit in Standard_NV8as_v4's 16 GB of GPU memory. VRAM is estimated from the parameter count at the selected quantization, so treat it as guidance rather than a guarantee.

Chat Models

FP16 (full precision)

Model	Creator	Intelligence	Total params	Active params	~VRAM
Qwen3.5 4B	Alibaba	20.1#191	4B	—	10 GB
MiniCPM5 1B	Openbmb	12.0#292	1B	—	2 GB
MiniCPM5 1B Non-Reasoning	Openbmb	11.7#300	1B	—	2 GB
7 more models fit on Standard_NV8as_v4 Unlock the full ranked list and FP8 / INT4 quantization with a CloudPrice subscription.

Similar Alternative VMs

Find Similar Instances In:

Microsoft Azure