Early access open
Know your AI model is ready for the GPU before you deploy it
Upload your model. We run it on real cloud GPUs and send you one clear report: does it work, are the answers right, and which GPU gives you the speed you need.
No setup · No GPU account needed · Your model stays private
Deploying a model shouldn't be a gamble
“It worked on my laptop.”
Then it fails, slows down or behaves differently on the server GPU, after you have paid for it.
“Which GPU do I actually need?”
Rent too big and you overpay every month. Rent too small and you miss the speed your users expect.
“Is the faster setting safe?”
Speed-ups can quietly change your model's answers. You want to know before your customers do.
“Setting up GPUs takes days.”
Drivers, versions and runtimes stand between you and a single trustworthy number.
How it works
Upload your model
Send us the model file. No code, no setup, no explanation of how it works.
We test it on real GPUs
We run it on the GPUs you are considering, the same hardware you would rent, and check every answer.
Get your report
A clear verdict, the speed on each GPU, and our recommendation, with anything to fix in plain words.
What your report tells you
Does it work?
Whether your model runs on each GPU, and if not, exactly why.
Are the answers right?
Every GPU result checked against your model's original answers.
How fast is it?
Response time for one request, and how many it can handle per second.
Which GPU should you use?
A side-by-side comparison of GPUs and a clear recommendation.
What to change
Simple advice when a setting is unsafe or a part of your model holds it back.
See a real report
These are real results from our first tests on an NVIDIA RTX 5090, using well-known public models. Your report looks just like this.
Sorts a photo into 1,000 categories.
| Setup | Answers | Response time | Requests per second |
|---|---|---|---|
| CPU only | Same as original | 4.3 ms | 352 |
| GPU, standard | Same as original | 1.0 ms | 13,751 |
| GPU, half precision | Same as original | 0.66 ms | 21,590 |
| GPU, optimizedRecommendedTensorRT, half precision | Same as original | 0.31 ms | 61,741 |
Response time: one request on its own. Requests per second: the most it handled with requests grouped together.
What the report found
- On a GPU this model answers 14× faster than on a CPU.
- With requests grouped together, one GPU handles about 176× the traffic of the CPU.
- Every setup gave the same answers as the original model, so the faster ones are safe to use.
Find the right GPU for your model
Your report compares every GPU you choose. The RTX 5090 numbers are measured; the others are estimates until our tests on them are complete.
ResNet-18 · Image classification
| GPU | Memory | Response time | Requests per second |
|---|---|---|---|
| RTX 5090 Measured | 32 GB | 0.31 ms | 62,000 |
| RTX 4090 Estimate | 24 GB | 0.38 ms | 40,000 |
| L4 Estimate | 24 GB | 0.68 ms | 14,000 |
| A100 80 GB Estimate | 80 GB | 0.36 ms | 59,000 |
| H100 Estimate | 80 GB | 0.27 ms | 117,000 |
MiniLM · Text embeddings
| GPU | Memory | Response time | Requests per second |
|---|---|---|---|
| RTX 5090 Measured | 32 GB | 0.58 ms | 75,000 |
| RTX 4090 Estimate | 24 GB | 0.73 ms | 49,000 |
| L4 Estimate | 24 GB | 1.3 ms | 16,000 |
| A100 80 GB Estimate | 80 GB | 0.69 ms | 71,000 |
| H100 Estimate | 80 GB | 0.51 ms | 142,000 |
Whisper tiny · Speech
| GPU | Memory | Response time | Requests per second |
|---|---|---|---|
| RTX 5090 Measured | 32 GB | 1.0 ms | 1,401 |
| RTX 4090 Estimate | 24 GB | 1.3 ms | 911 |
| L4 Estimate | 24 GB | 2.3 ms | 308 |
| A100 80 GB Estimate | 80 GB | 1.2 ms | 1,331 |
| H100 Estimate | 80 GB | 0.88 ms | 2,662 |
Tiny YOLOv2 · Object detection
| GPU | Memory | Response time | Requests per second |
|---|---|---|---|
| RTX 5090 Measured | 32 GB | 0.32 ms | 21,000 |
| RTX 4090 Estimate | 24 GB | 0.40 ms | 14,000 |
| L4 Estimate | 24 GB | 0.70 ms | 4,574 |
| A100 80 GB Estimate | 80 GB | 0.37 ms | 20,000 |
| H100 Estimate | 80 GB | 0.28 ms | 40,000 |
Built for teams putting AI into production
AI startups
Launch your AI feature knowing it will hold up, without a GPU expert on the team.
ML engineers
Get production numbers for a new model version in one step, instead of a week of setup.
Teams choosing hardware
Pick a GPU based on how your own model performs, not on spec sheets.
Why GridPup
- Real GPUs, not simulations. Every number comes from your model running on the actual hardware.
- Many GPUs in one place. GridPup already tracks cloud GPUs across providers, so you can test on the ones you would really rent.
- Results you can trust. Our speed measurements match NVIDIA's own benchmark tools on the same GPU.
- Private by default. Your model is used for your test only.
Questions
What models can I test?
We start with ONNX, the format PyTorch, TensorFlow, scikit-learn and most other tools can export to: image, text, speech, recommendation and tabular models. More formats will follow.
Do I need my own GPU or cloud account?
No. We provide the GPUs, set them up and run everything. You upload a model and get a report.
Is my model kept private?
Yes. Your model is used only for your test and is not shared with anyone.
Do I have to tell you how my model works?
No. We work out what your model needs as input. If you have a few real examples, you can add them for an even more realistic test.
When can I start, and what will it cost?
We are opening early access now. Plans will be announced at launch, and early-access members hear first.
Get early access
Be among the first to test your models on real GPUs. Tell us a little about what you want to test and we will reach out when your spot is ready.
Questions? Contact us.