OctaneRender for Blender on Multiple GPUs: Octane Server, Devices and Versions
Octane in Blender runs as an add-on plus a separate OctaneRender Server. Check these before a multi-GPU render:
- Versions: add-on 2026.5 – 31.12 is only for Blender 5.2.2; install the matching Octane Server
- Devices: choose GPUs in Render Properties > Octane Server > Device Preferences
- Not Cycles: Blender’s own GPU preferences apply to Cycles, not to Octane
- Memory: VRAM is not combined; the smallest active GPU sets the in-core limit
- License: OTOY’s Studio+ plan has no limit on the number of GPUs
OctaneRender for Blender changed shape in 2026. OTOY stopped making its own custom Blender builds and moved to an add-on that runs in the standard Blender from blender.org, together with a separate program, the OctaneRender Server, which does the rendering. That split decides where you pick GPUs, which versions have to match and how animation frames reach the GPUs.
This guide explains how the add-on and server work together on a multi-GPU machine, where to select devices, how memory works across cards, and what to install on a render machine.
Checked against OTOY’s OctaneRender for Blender add-on manual and release posts (2026.3 – 31.9 to 2026.5 – 31.12), and OTOY device documentation, November 2026.
1. How does OctaneRender for Blender work?
The installation has two parts: the OctaneRender Server and the OctaneRender for Blender add-on. OTOY’s install steps start the server first, then install the add-on in Blender from Preferences, then create a scene from File > New > OctaneRender Startup. When you render, the add-on sends the scene to the server, the server renders it on the GPUs and the image comes back to Blender.
The first stable add-on, 2026.3 – 31.9, removed the need for a custom Blender build. The current stable release at the time of writing is 2026.5 – 31.12, published on OTOY’s forum on 9 October 2026. OTOY states it is compatible only with Blender 5.2.2; earlier Blender 5.x and 4.x versions are not supported. Each add-on release targets one Blender version, so updating Blender usually means waiting for, or installing, the matching add-on.
For installation and the free Prime tier, see our guide to installing OctaneRender for Blender.
2. Where do you choose GPUs for Octane in Blender?
Not in Blender’s Preferences. The GPU list under Preferences > System belongs to Cycles; as users on OTOY’s forum point out, those settings do not apply to Octane. Octane’s devices are set on the server side: in Render Properties, the Octane Server rollout has a Device Preferences button that shows the GPU resources and settings, alongside Network Preferences and the license sign-in.
Device Preferences works like the Octane device settings in other hosts: tick each GPU that should render and set its priority. On a workstation, give the card that drives your monitor a lower priority; on a render machine, enable every GPU at High priority. OTOY’s device documentation recommends not mixing a weaker card into a set of identical GPUs.
Memory across several GPUs
Speed adds up across GPUs, memory does not. OTOY’s device documentation states that every GPU needs a copy of the scene assets it processes, so the usable memory is set by the card with the least VRAM. Four 32 GB cards render a scene that fits in 32 GB, four times faster than one card would, but they do not give 128 GB. Out-of-core can place textures and geometry in system RAM at a speed cost. Our Octane multi-GPU guide covers devices and memory in more depth; the rules are the same in Blender.
3. Versions that must match on a render machine
Because Blender, the add-on and the server are separate pieces, all of them have to agree. OTOY ties each add-on to one Blender version and one Octane version, and its network rendering documentation for Blender is explicit that a render node must run the same version as the server it helps.
In practice: install Blender 5.2.2 from blender.org, the 2026.5 – 31.12 add-on and the 2026.5 OctaneRender Server, and start the server before Blender. A scene saved in an older add-on version may need checking in the new one; render a test frame and compare it with your reference before the full range.
The server download is offered for Windows, Linux and macOS, and in Studio and Prime editions. OTOY’s subscription page states that Studio+ has no limit on the number of GPUs, so use a Studio+ account on multi-GPU machines.
4. Rendering animations with Octane in Blender
For an animation, the add-on hands each frame’s scene to the server before the GPUs start sampling. On OTOY’s forum, Blender users have described the status messages that appear during this step and the preparation time it adds to each frame in heavy scenes. As in Cinema 4D, more GPUs shorten only the sampling part of each frame, not the preparation.
Two habits help on long sequences. Measure a few consecutive frames and note how much of each frame is preparation and how much is rendering. And if preparation dominates, split the frame range across several machines rather than adding GPUs to one. Our guide to Octane sequence QA lists the frames worth testing; the checks apply to Blender output too.
5. When Blender cannot reach Octane Server, and rendering without the interface
Most “Octane is not available” problems on a new machine come from the order of installation, not from the GPUs. OTOY’s installation guide gives the rules:
- Server first: run
OctaneServer.exebefore opening Blender; it must stay active for Octane to work. - One server: close any running OctaneServer before starting it again.
- Matching numbers: the server and add-on versions must match, and the add-on must be built for your Blender version.
- Zipped add-on: install the add-on zip with Install from Disk without extracting it.
- Sign-in: sign in with your OTOY account if prompted, so the licensing system can fetch your license.
Once Octane renders a frame in the interface, the Octane Server rollout in Render Properties has Show Octane Log for tracking what the server does. For long sequences you can then render without the Blender window, using Blender’s own command line: -b runs in the background, -s and -e set the first and last frame, and -a renders the animation with the scene’s output settings.
Test with a short range such as -s 1 -e 2 first and check the output folder, because a Blender user on OTOY’s forum reported command-line renders that quit without rendering until an Octane update fixed it. Splitting the range across machines suits scenes where per-frame preparation, not sampling, takes most of the time.
Sources: OTOY: OctaneRender for Blender Installation Guide, OTOY: Octane Server rollout, Blender manual: rendering from the command line, OTOY forum: batch render from the command line.
6. SaaS or IaaS render farm for Octane in Blender?
Which farm type fits an Octane for Blender job
The add-on is tied to one Blender version, so version control decides most cases.
| Situation | Usually a better fit | Why |
|---|---|---|
| Farm runs your exact Blender and add-on versions | SaaS (managed farm) | Submit the .blend and render |
| You need a specific Blender + add-on pair | IaaS (remote workstation) | Install the exact versions yourself |
| Device or memory setup still uncertain | IaaS | Check Device Preferences and per-GPU memory on the machine |
| Long animation with heavy preparation | Either, split across machines | Each machine prepares its own frames |
| Short one-off render | SaaS | No setup time |
Rule of thumb: with Octane for Blender, the version pair matters more than anything else. If a farm cannot run your exact Blender and add-on versions, use a machine where you install them yourself.
Rendering Octane for Blender on an iRender server
On iRender you rent a whole Windows workstation with full administrator access, so you set up the same Blender and Octane versions you use at home:
- Upload the .blend file and textures through iRender Drive.
- Start a 2x, 4x or 8x RTX 5090 server (32 GB per card).
- Install Blender 5.2.2, the OctaneRender Server 2026.5 and the 2026.5 – 31.12 add-on; start the server and sign in with your own OTOY account.
- In Render Properties > Octane Server > Device Preferences, enable every GPU and render a test frame.
- Render the animation; your installed versions stay on your saved server image for the next session.
Octane for Blender checklist
Use this as a final check before you start the full range on any machine.
Frequently asked questions
1. Does OctaneRender for Blender use multiple GPUs?
Yes. The OctaneRender Server renders each frame on every GPU enabled in Device Preferences, and OTOY describes the speed gain as roughly linear with more GPUs.
2. Where do I select GPUs for Octane in Blender?
In Render Properties, open the Octane Server rollout and click Device Preferences. Blender’s own GPU settings in Preferences apply to Cycles, not to Octane.
3. Which Blender version does the Octane add-on support?
Each add-on release targets one Blender version. The stable 2026.5 – 31.12 release from October 2026 supports only Blender 5.2.2.
4. Do I still need a custom OctaneBlender build?
No. Since the stable 2026.3 – 31.9 add-on, Octane runs in the standard Blender from blender.org, and OTOY has stopped making new custom builds.
5. Does Octane combine VRAM across GPUs in Blender?
No. Each GPU needs a copy of the scene it renders, so the smallest active card sets the in-core limit. Out-of-core can use system RAM at a speed cost.
6. Do Octane network render nodes need the same version?
Yes. OTOY’s Blender documentation states that a render node must run the same version as the OctaneRender Server it helps.
7. Why is my Octane animation in Blender slow between frames?
Each frame’s scene goes from Blender to the server before rendering starts. More GPUs shorten only the rendering part; split long sequences across machines if preparation dominates.
Render your Blender + Octane scenes on iRender
iRender is an IaaS GPU render farm: you rent a bare-metal server with 1, 2, 4 or 8 NVIDIA RTX 4090 or RTX 5090 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. You install the Blender and OctaneRender for Blender versions you use and sign in with your own OTOY account (OTOY’s Studio+ plan has no GPU limit for multi-GPU servers). Your server image is saved between sessions.
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, so they keep their boost clocks. Elite support is available 7 days a week by hotline, live chat, Skype, WhatsApp and email.
This quick start shows how to start Blender on an iRender server:
For Octane speed on these servers: on 4x RTX 5090, a 300-frame Octane animation from Cinema 4D rendered in 1 hour 18 minutes. The second clip runs the same kind of scene on 4x RTX 5090 and 4x RTX 4090:
New users get a 100% bonus on the first top-up (minimum $50). See plans on the Octane render farm page, download the iRender GPU app, or read Octane multi-GPU rendering.
Sources: OTOY: OctaneRender v2026 for Blender add-on manual, OTOY: Octane Server rollout (Blender), OTOY: GPU settings (Blender), OTOY: Blender add-on 2026.5 – 31.12 release, OTOY: Blender add-on 2026.3 – 31.9, first stable, OTOY forum: end of custom OctaneBlender builds, OTOY: network rendering (Blender), OTOY: Devices tab, VRAM across GPUs, OTOY: OctaneRender Subscriptions. Checked November 2026.




