The problem D5 users actually have.
D5 is a real-time renderer, which is why weak hardware hurts it so much. You use it because you move the camera and the image resolves while you watch. On an underpowered card that loop breaks: the viewport stutters, you stop exploring options, and a 4K still becomes something you set running and walk away from.
The usual answer is to buy a workstation. A machine that renders comfortably costs roughly $6,000. You buy it once and it sits idle most of the week, because visualisation work comes in bursts. You need serious hardware for the four days before a client presentation and almost none for the three weeks after it.
Renting flips that around: you have the machine for those four days and don’t pay for the three weeks.
The machine.
NVIDIA RTX A4000 16 GBEntryIntro hours apply
- Ampere
- 6,144 CUDA cores
- 16 GB VRAM
- 16 vCPU
- 64 GB RAM
$1.49
per GPU-hour, no GST
NVIDIA RTX A5000 24 GBMore memoryIntro hours apply
- Ampere
- 8,192 CUDA cores
- 24 GB VRAM
- 16 vCPU
- 64 GB RAM
$2.29
per GPU-hour, no GST
RTX 5080Fast renderingIntro hours apply
- Blackwell
- 10,752 CUDA cores
- 16 GB VRAM
- 16 vCPU
- 64 GB RAM
$3.49
per GPU-hour, no GST
NVIDIA RTX 6000 Ada 48 GBBig scenes
- Ada
- 18,176 CUDA cores
- 48 GB VRAM
- 16 vCPU
- 128 GB RAM
$5.29
per GPU-hour, no GST
- Disk
- 1 TB NVMe
- OS
- Windows 11
- Access
- Coreframe Connect app, streamed at 4K at 60 fps
- Display driver
- WDDM, which D5 needs for real-time viewport work
- D5 Render
- Preinstalled; sign in with your own D5 account
- Project storage
- Persistent, separate from the machine
- Location
- Bengaluru, India
Each node has one GPU and one customer at a time. You don’t share the card, and you don’t get a virtualised slice of it. The number we’d compare is VRAM: 16 to 48 GB here, while many laptops and entry desktops sold for design work have 6 to 8 GB. VRAM decides whether a scene loads at full texture resolution or has to be cut down. We publish a render time only once we have measured it on this hardware ourselves; the table below says which ones we have.
How D5 Render runs here: measured, not quoted.
We publish only figures we have measured on our own workstation, and we have not published D5 Render timings yet. Each row says so. Ask and we will run your scene, or a reference one, and send the numbers.
| What was measured | Result |
|---|---|
| Time to open a mid-size D5 Render project from the project drive | Measured on request |
| D5 Render 4K still, default quality, on our reference scene | Measured on request |
| Peak VRAM used by that reference scene | Measured on request |
| Workstation ready after Launch (sign-in to desktop) | Measured on request |
Want a number before you top up? Ask for a measured run. We will tell you the GPU, the scene and the settings with every figure.
How a session works.
- 1Sign in and start a workstation. It comes up in about two minutes. Provisioning is free, and billing starts when the stream does.
- 2Your project files are already there, on persistent storage that survives between sessions. Nothing has to be copied onto the machine each time.
- 3Open D5, sign in with your own account, and work. Twinmotion, Blender and Unreal Engine are on the same desktop if you need them.
- 4Render at full resolution on the workstation. The finished image is a file you download. It doesn't come through the stream.
- 5Close the session. Billing stops, the machine is wiped back to a clean image, and your files stay where they were.
The full setup walkthrough covers sign-up and installing Coreframe Connect.
The revision that happens in the room.
Studios feel this most in client meetings. A client asks for the kitchen in a different finish, and the normal answer is to go back to the office, re-render overnight and book a second meeting. With the workstation reachable from a laptop, you can make the change where you’re sitting and put the result on the client’s own screen before the meeting ends.
Two caveats. Test the connection at the client’s office first, or tether, because a demo that stalls in front of the client is worse than no demo. And the high-resolution still is a file rendered on the workstation. You won’t see it as a live stream at that resolution.
What D5 Render costs per hour here.
$1.49/hr, in US dollars with no GST (export of services). Billed per minute from when you connect, not per whole hour. Preparing the workstation and uploads are free. If you don't connect within 10 minutes of your workstation being ready, it shuts down and those 10 minutes are charged. If a connection fails on our side, nothing is charged.
We read that price from the same rate card billing charges from, so the figure on this page can’t drift from what you’re actually charged. An invoice in US dollars is issued on every recharge. Full pricing and monthly plans.
Questions.
Can I run D5 Render on a cloud workstation?
Yes. Coreframe gives you a full Windows 11 desktop with a dedicated NVIDIA RTX GPU, and D5 Render is already installed on it. You sign in with your own D5 account, open your project and work in the live viewport the same way you would on a local machine. It's the full desktop application. You reach it through the Coreframe Connect app rather than a remote desktop, and the stream runs at 4K at 60 fps.
Is D5 Render included, or do I need my own licence?
The application is installed for you, so you never spend paid minutes downloading it. The account is yours: you sign in with your own D5 credentials, and whatever plan you hold with D5 is what applies. We supply the machine. The licence is yours.
Is this the same as a render farm?
No. A render farm takes a finished scene, renders frames on machines you never see, and sends images back. Coreframe hands you the machine itself. You can move the camera, change a material, relight the scene and re-render in the same session. That back-and-forth is how people use D5, and it's why you'd pick a real-time renderer in the first place.
How much VRAM does D5 Render need?
VRAM is the ceiling on scene size in any GPU renderer: geometry, textures and lightmaps all have to fit on the card. A Coreframe workstation has 16 to 48 GB of VRAM depending on the GPU you pick. Many laptops and entry desktops sold for design work ship with 6 to 8 GB, which is what forces people to cut texture resolution or split a scene. We can't tell you how your particular scene will behave, so run it in a session and watch the VRAM meter yourself.
What happens to my D5 project files when the session ends?
Every session starts from an identical clean machine, so anything installed during a session is gone when it ends. Your project files don't live on that machine. They sit on persistent storage that carries over, so a session you open on Friday picks up where Monday's stopped. The clean reset is how we make sure no trace of another customer's work, or yours, is left behind.
Can I use D5 Render with Revit, SketchUp, Rhino or 3ds Max on the same machine?
Yes. It's an ordinary Windows desktop, so the D5 plugins and the host applications sit side by side. Blender and Twinmotion are preinstalled alongside D5. Revit, SketchUp, Rhino, 3ds Max and similar licensed applications run on the machine but need your own licence, because those vendors require your own account even to download the installer.
What internet speed do I need to use a Coreframe workstation?
You're streaming a desktop, so a steady connection matters more than a fast one. 4K at 60 fps looks its best on about 50 Mbps and still holds up at 35 Mbps. 1080p is comfortable on 20 Mbps. Rendering happens on the workstation in Bengaluru, and a finished 4K or 8K still is a file you download. It isn't streamed to you frame by frame.
How much does it cost to render in D5 on Coreframe?
Ad-hoc sessions are $1.49 per GPU-hour, billed per minute from when you connect, not per whole hour. Preparing the workstation and uploads are free. If you don't connect within 10 minutes of your workstation being ready, it shuts down and those 10 minutes are charged. If a connection fails on our side, nothing is charged. Committed monthly plans start at $70.99/month and bill extra hours below the ad-hoc rate at every tier: $1.99/hr on Small Team, $0.99/hr on Medium Team. Prices in US dollars carry no GST (export of services), and the invoice is in US dollars.
Is there an offer for new customers?
Yes. Your first top-up of $20 (no GST) buys 8 hours on an RTX 5080 (or about 18 hours on the RTX A4000, 12 on the RTX A5000), to use within 30 days, one per customer. You sign up with an email and a WhatsApp code to your mobile, with no ID documents, and international cards work. We won't say what you can finish in that time, because it depends entirely on your scene.
Test it with your own scene.
Take the D5 project that ties up your machine all afternoon and run it on ours. You’ll learn more from that than from any number we could put on this page.
Get started