November 9, 2026 Kath Nguyen

V-Ray GPU for Maya on Multi-GPU Machines: CUDA vs RTX, Memory and Device Setup

V-Ray GPUV-Ray GPU in Maya on several GPUs
From Chaos documentation

V-Ray GPU behaves differently from V-Ray CPU. Check these before a long render:

  • Engine: V-Ray GPU uses a separate code base and supports a subset of V-Ray features
  • Platform: choose CUDA or RTX in the V-Ray render settings
  • Memory: GPU memory use is capped by the device with the least memory
  • Devices: pick render GPUs with the Select Devices tool
  • Out of memory: enable GPU texture resizing with a size limit

V-Ray for Maya can render on the CPU or on NVIDIA GPUs, and on a machine with several GPUs the GPU engine is usually the faster choice for look development and animation. It is not a switch you flip without checking: V-Ray GPU is a separate engine with its own feature set, its own memory rules and two platforms to choose from.

This guide covers how V-Ray GPU differs from V-Ray CPU in Maya, when to pick CUDA or RTX, why mixed GPUs can waste memory, how to select devices, and how to deal with out-of-memory errors before a long batch.

Checked against Chaos’s V-Ray for Maya documentation (FAQ), Chaos’s RTX announcement and CG Channel’s release coverage, November 2026.

1. How V-Ray GPU differs from V-Ray CPU in Maya

Chaos states that V-Ray GPU uses a different code base from the CPU engine, optimized for GPU hardware, and supports a subset of V-Ray’s features, listed on its Supported Features page. With V-Ray GPU you choose the CUDA or RTX platform in the V-Ray tab of Render Settings (Chaos: V-Ray for Maya FAQ). Chaos introduced the RTX engine as an option that uses the RT Cores in RTX GPUs (Chaos: V-Ray GPU adds RTX support), and CG Channel reported that RTX mode supports V-Ray GPU features except hybrid CPU and GPU rendering (CG Channel).

iRender
V-Ray GPU in Maya: Memory and Engine Choice
Schematic
A · Mixed cards: memory follows the smallest4 GB card8 GB card12 GB cardusable: 4 GB on every cardChaos example: 4, 8 and 12 GB cards use only 4 GBB · CUDA or RTX engineCUDANVIDIA GPUs, and the CPU can be addedas an extra device (hybrid)the CPU works on the same frame as the GPUsRTXuses the RT cores on RTX cardsGPUs only, no hybrid CPU modetest both on your heaviest frame
memory V-Ray can usecap set by smallest cardCUDARTX
Schematic · based on Chaos docs and CG Channel

Panel A matters most on machines with different cards: Chaos gives the example that with 4, 8 and 12 GB cards, only 4 GB is used (Chaos). On a machine with matching cards, every GPU can use its full memory.

2. Managing GPU memory in V-Ray for Maya

When V-Ray GPU reports “Failed to allocate memory for buffer”, Chaos says to enable GPU texture resizing and set a texture size limit (Chaos). V-Ray 6 for Maya Update 2 also added a Compressed Textures mode to reduce memory use on the GPU (CG Channel: V-Ray 6 for Maya Update 2). Before turning either on, find which textures are heaviest:

iRender
Find the heaviest textures in the scene
Python
1# List the 15 largest texture files used by file nodes in the scene
2# Run in Maya's Script Editor (Python tab) before a V-Ray GPU render
3import os
4import maya.cmds as cmds
5rows = []
6for node in cmds.ls(type="file"):
7 path = cmds.getAttr(node + ".fileTextureName")
8 if path and os.path.isfile(path):
9 rows.append((os.path.getsize(path) / 1048576.0, node, path))
10rows.sort(reverse=True)
11print("total MB on disk:", round(sum(r[0] for r in rows), 1))
12for mb, node, path in rows[:15]:
13 print(round(mb, 1), "MB", node, path)
Replace the paths, names and frame range with your ownAutodesk Maya Python commands

The script lists file sizes on disk, not the memory V-Ray will use after loading, but it shows quickly which maps to resize or compress first. Re-render the heaviest frame after each change and compare the result.

3. How to set up V-Ray GPU in Maya (step by step)

Chaos’s Select Devices for V-Ray GPU rendering tool lists the devices that can render; if the list is empty, update the video driver. With several GPUs you can remove the one that drives the monitor or viewport; with one GPU you can set it to Low Priority to keep the interface responsive (Chaos).

Settings checklist (copy or follow line by line):

Settings checklist
Render Settings › V-Ray
Engine → V-Ray GPU
Platform → RTX on RTX cards; CUDA if you need hybrid
Devices → all render GPUs ticked in Select Devices
Memory
Cards → same model, so the smallest is not a bottleneck
Textures → resize or compress if memory runs out
Heaviest frame → memory read in the log
Scene
Features → checked against the GPU Supported Features list
Look → compared with CPU frames if those were approved

4. Common V-Ray GPU problems in Maya

Most V-Ray GPU problems come from memory, device selection or a feature that the GPU engine handles differently. The table maps each one to its fix.

TroubleshootV-Ray GPU for Maya

When V-Ray GPU does not render as expected

Start from what you see.

What you seeLikely causeWhat to change
No devices listed for GPU renderingDriver not installed or too oldUpdate the NVIDIA driver, then run Select Devices again
“Failed to allocate memory for buffer”Scene needs more GPU memory than the smallest card hasEnable GPU texture resizing with a size limit, or use cards with more memory
Big card shows only part of its memory usedMemory is capped by the smallest device in the listUntick the smallest card or use matching cards
A look differs from V-Ray CPUV-Ray GPU is a separate code base with a subset of featuresCheck the feature against the GPU Supported Features list
Hybrid CPU option missingRTX platform selectedSwitch to CUDA if you need the CPU as a device
Viewport slow while rendering on one GPUDisplay GPU also rendersSet it to Low Priority or remove it from the device list

For how V-Ray GPU compares with other engines that use several GPUs in Maya, see the best multi-GPU render engines for Maya 2026.

5. SaaS or IaaS render farm for V-Ray GPU in Maya?

CompareNeutral guide

Which farm type fits V-Ray GPU work

Both farm types can render V-Ray scenes; the difference is licensing and who chooses devices and settings.

SituationUsually a better fitWhy
Final frames, farm supports your V-Ray versionSaaSUpload the scene and collect frames
Your own V-Ray license and versionIaaSYou install V-Ray and sign in on the render machine
Memory or device settings still being tunedIaaSYou change texture limits and devices and re-test
CPU and GPU frames must be comparedIaaSRender both on the same machine

Rule of thumb: tune memory and devices on a machine you can open, then render where licensing fits.

Rendering V-Ray GPU for Maya on an iRender server

iRender
How iRender works: your computer controls your own multi-GPU server
IaaS render farm
Your PC or MacNo powerful GPU neededRemote desktopLive screen1:1 like your PCYOUR OWN GPU SERVERbare-metal · not sharedGPUGPUGPUGPUGPUGPUGPUGPU1, 2, 4 or 8 × RTX 5090 (32 GB) or RTX 4090 (24 GB)Threadripper PRO · 256 GB ECC RAM · 2 TB NVMe SSDFull admin access · your setup is saved between sessions1Register, top upFirst $50+ top-up doubled2Upload your projectiRender Drive, free3Start and connectRemote desktop, any OS4Render, shut downBilling stops, per minute✓ Your workstation, just more GPUs: no queue to wait, no farm plugin, no job submission, no settings changed
Your mouse and keyboardServer screen streamed back
Schematic, not to scale

An iRender server is a full Windows machine with 32 GB RTX 5090 or 24 GB RTX 4090 GPUs; all cards in a server are the same model:

  1. Upload the project to iRender Drive before you start a server; uploading is free and does not use rental time.
  2. Start a server, install your Maya version and V-Ray for Maya, and sign in to your own V-Ray license; they stay in your image.
  3. Select all GPUs in Select Devices, render the heaviest frame and check memory.
  4. Render the full range, then shut down. Multi-GPU servers bill per minute; 1-GPU servers have a one-hour minimum.

V-Ray GPU checklist

Tick each item before you start the full range.

ChecklistV-Ray GPU checklist
0 / 10 done
Tip: tick each item as you set up the job; the ticks reset when the page reloads.

Frequently asked questions

1. Is V-Ray GPU the same engine as V-Ray CPU?

No. Chaos states V-Ray GPU uses a different code base optimized for GPUs and supports a subset of V-Ray features, listed on its Supported Features page.

2. Should I use CUDA or RTX in V-Ray GPU?

RTX uses the RT Cores on RTX cards; CUDA can also add the CPU as a device in hybrid mode, which RTX mode does not support. Test both on your heaviest frame. For Arnold GPU and Redshift in Maya, see Arnold GPU vs Redshift in Maya 2026.

3. Why does V-Ray GPU not use all the memory on my big card?

GPU memory use is capped by the device with the least memory. Chaos’s example: with 4, 8 and 12 GB cards, only 4 GB is used.

4. How do I fix “Failed to allocate memory for buffer” in V-Ray GPU?

Chaos suggests enabling GPU texture resizing and setting a texture size limit. V-Ray 6 for Maya Update 2 also added a Compressed Textures mode to cut memory use.

5. How do I choose which GPUs V-Ray uses in Maya?

Use the Select Devices for V-Ray GPU rendering tool. You can remove the GPU that drives the display, or set a single GPU to Low Priority to keep Maya responsive.

6. Why is my GPU device list empty?

Chaos says to update the video driver when no devices are listed.

7. Does V-Ray GPU work across several machines?

Chaos notes you can speed up rendering with distributed rendering across systems that have CUDA-capable devices.

8. Can I render V-Ray GPU for Maya on iRender?

Yes. V-Ray is installed by you with your own license on a full Windows server; every server uses matching GPUs, so memory is not limited by a smaller card.

Render your Maya projects on iRender

iRender is an IaaS GPU render farm: you rent a bare-metal server with 1, 2, 4 or 8 NVIDIA RTX 4090 (24 GB) or RTX 5090 (32 GB) GPUs, an AMD Ryzen™ Threadripper™ PRO CPU, 256 GB ECC RAM and a 2 TB NVMe SSD, connect by remote desktop and work on it like your own workstation, with full administrator access. Your server image is saved between sessions, so your plugins and settings stay installed. Maya is not pre-installed: you install the Maya version your scene uses with your own Autodesk license, and it stays in your image. Prime accounts get Redshift licenses pre-activated on every server.

Multi-GPU servers are billed per minute (1-GPU servers have a one-hour minimum), and sessions of 3 hours or more get 10% off. Our 4-GPU and 8-GPU servers are liquid-cooled; in our own measurements the GPUs hold 60–70°C under sustained render load. Elite support is available 7 days a week by hotline, live chat, Skype, WhatsApp and email.

This quick start shows how to start a server and connect to it:

A full Maya and Redshift session we recorded on an 8x RTX 4090 server:

New users get a 100% bonus on the first top-up (minimum $50). See plans on the Maya render farm for Redshift page or download the iRender GPU app.

Accelerate your production pipeline with high-performance IaaS render farm services and dedicated bare-metal workstations powered by up to 8x NVIDIA GeForce RTX 5090 (32GB GDDR7), high-frequency AMD Ryzen™ Threadripper™ PRO processors, 256GB ECC RAM, and dedicated 2TB NVMe SSD storage tailored for your pipeline:
, , , , , , ,

Kath Nguyen

Kath Nguyen works in customer support at iRender, helping artists move Blender and Cinema 4D projects to multi-GPU servers. She has written more than 140 articles for the iRender blog.
Contact

iRENDER GPU

Render Farm Pricing
Quick Start
Multi-GPU Benchmarks
iRender GPU App
Blog
Documentation
The Render Farm
The Company
Terms & Conditions

INTEGRATIONS

Autodesk Maya
Blender
Cinema 4D
Houdini
Karma XPU
Lumion
KeyShot
Unreal Engine
Twinmotion
Redshift
Octane
And many more…

iRENDER TEAM

Elite support 7 days a week
SATURDAY – SUNDAY: 6:00 AM – 11:59 PM
(UTC+7)
Hotline: (+84) 912-785-500
Skype: iRender Support
Email: [email protected]
Address 1: 68 Circular Road #02-01, 049422, Singapore.
Address 2: No.22 Thanh Cong Street, Hanoi, Vietnam.

Contact
이 페이지는 한국어로도 제공됩니다. 한국어로 보기 →