-
×InformationNeed Windows 11 help?Check documents on compatibility, FAQs, upgrade information and available fixes.
Windows 11 Support Center. -
-
×InformationNeed Windows 11 help?Check documents on compatibility, FAQs, upgrade information and available fixes.
Windows 11 Support Center. -
- HP Community
- Notebooks
- AI PCs
- Re: Can I upgrade my HP Desktop to add a NVIDIA Tesla P100 1...

Create an account on the HP Community to personalize your profile and ask a question
06-26-2025 09:17 PM
Can I upgrade my HP Desktop to add a NVIDIA Tesla P100 16GB HBM2 PCIe Graphics Card? Is this motherboard capable of running two graphics cards? Can I add a power supply large enough to it? Is my ram fast enough for this card?
Solved! Go to Solution.
Accepted Solutions
06-28-2025 10:55 AM
@MZinzow, Thank you for your response.
Thanks for the clarification! Since your goal is to run larger AI models (like text-to-video in Amuse 3.x), maximizing VRAM is a smart move. Let’s break down your options and what’s realistically compatible with your HP Pavilion Gaming Desktop 690-0067c.
Key Considerations for VRAM-Heavy GPUs:
- Motherboard: You have one PCIe x16 slot (Gen 3.0).
- Case Size: Mid-tower, but space is limited.
- Power Supply: Stock 400W – must be upgraded for high-end GPUs.
- Cooling & Airflow: Limited, so avoid workstation/datacenter GPUs like Tesla or Quadro unless you mod the case.
Compatible GPUs with More than 12GB VRAM:
- NVIDIA RTX 3080
10–12 GB
750W
- NVIDIA RTX 3090
24 GB
850W
- NVIDIA RTX 4070 Ti Super
16 GB
700W
- AMD RX 7900 XT/XTX
20–24 GB
750W
- NVIDIA RTX A4000 (Workstation)
16 GB
550W
- NVIDIA RTX A5000
24 GB
750W
Recommended Upgrade Path:
- GPU: Consider the RTX 4070 Ti Super (16GB) or RTX A4000 (16GB) – both offer great AI performance and are more likely to fit.
- Power Supply: Upgrade to a 650W–750W PSU from a reliable brand (Corsair RMx, EVGA GQ, Seasonic Focus).
- Case Fit: Measure your case’s GPU clearance (~10.5–11 inches max). Some high-end GPUs are 12–13 inches long.
AI Model Compatibility:
- VRAM is critical for loading large models, especially for video generation.
- RTX 4070 Ti Super supports Tensor cores and CUDA, which are essential for frameworks like PyTorch and TensorFlow.
- RTX A4000 is a great workstation card with 16GB VRAM and lower power draw (~140W), ideal for tight builds.
I hope this helps!
If my response resolves your issue, please click “Accepted Solution” to help others find the answer. Also, don’t forget to click the “Yes” button to say thanks!
Take care and have a great day.
Max3Aj
HP Support
06-28-2025 08:10 AM
@MZinzow, Welcome to HP Support Community.
Thank you for posting your query, I will be glad to help you.
Let's walk through your upgrade questions for the HP Pavilion Gaming Desktop 690-0067c and the possibility of installing a NVIDIA Tesla P100 16GB HBM2 PCIe GPU.
System Overview:
Your desktop comes with:
- Motherboard: HP "Sunflower" (Micro-ATX)
- CPU: AMD Ryzen 7 1700
- RAM: 16 GB DDR4-2666
- GPU (stock): AMD Radeon RX 550
- Power Supply: 400W (standard)
Tesla P100 Compatibility Concerns:
Form Factor & Cooling:
- The Tesla P100 is a data center GPU designed for passive cooling and requires special airflow not available in standard consumer cases.
- Your Pavilion case is compact and not designed for high-thermal, passively cooled cards.
Power Supply:
- Tesla P100 requires 250W+ and a 6-pin or 8-pin PCIe power connector.
- Your stock PSU is 400W, which is insufficient. Even with an upgrade, the motherboard and case may not support the physical and thermal demands of the Tesla card.
Dual GPU Support:
- The "Sunflower" motherboard has only one PCIe x16 slot, so dual GPU setup is not possible.
RAM Compatibility:
- While your DDR4-2666 RAM is decent, the Tesla P100 doesn’t rely on system RAM for performance. However, system bottlenecks may still occur due to CPU and motherboard limitations.
Recommendations:
If you're looking to upgrade for gaming or GPU compute, consider:
- NVIDIA RTX 3060/3070 or AMD RX 6700 XT – these are consumer-grade, actively cooled, and compatible with upgraded PSUs.
- Upgrade your PSU to at least 600W from a reputable brand (e.g., Corsair, EVGA).
I hope this helps.
Take care and have a good day.
Please click “Accepted Solution” if you feel my post solved your issue, it will help others find the solution. Select "Yes" on the bottom left to say “Thanks” for helping!
Max3Aj
HP Support
06-28-2025 09:47 AM
Thank you. My goal is to get the most VRAM in this system I can for larger AI models.
My current system won't run even a single text to video model in Amuse 3.x.
Are there any cards with more than the 12Gb available with the RX 6700 XT that will work in my system? Does the RX 6700 XT also require a power supply upgrade?
06-28-2025 10:55 AM
@MZinzow, Thank you for your response.
Thanks for the clarification! Since your goal is to run larger AI models (like text-to-video in Amuse 3.x), maximizing VRAM is a smart move. Let’s break down your options and what’s realistically compatible with your HP Pavilion Gaming Desktop 690-0067c.
Key Considerations for VRAM-Heavy GPUs:
- Motherboard: You have one PCIe x16 slot (Gen 3.0).
- Case Size: Mid-tower, but space is limited.
- Power Supply: Stock 400W – must be upgraded for high-end GPUs.
- Cooling & Airflow: Limited, so avoid workstation/datacenter GPUs like Tesla or Quadro unless you mod the case.
Compatible GPUs with More than 12GB VRAM:
- NVIDIA RTX 3080
10–12 GB
750W
- NVIDIA RTX 3090
24 GB
850W
- NVIDIA RTX 4070 Ti Super
16 GB
700W
- AMD RX 7900 XT/XTX
20–24 GB
750W
- NVIDIA RTX A4000 (Workstation)
16 GB
550W
- NVIDIA RTX A5000
24 GB
750W
Recommended Upgrade Path:
- GPU: Consider the RTX 4070 Ti Super (16GB) or RTX A4000 (16GB) – both offer great AI performance and are more likely to fit.
- Power Supply: Upgrade to a 650W–750W PSU from a reliable brand (Corsair RMx, EVGA GQ, Seasonic Focus).
- Case Fit: Measure your case’s GPU clearance (~10.5–11 inches max). Some high-end GPUs are 12–13 inches long.
AI Model Compatibility:
- VRAM is critical for loading large models, especially for video generation.
- RTX 4070 Ti Super supports Tensor cores and CUDA, which are essential for frameworks like PyTorch and TensorFlow.
- RTX A4000 is a great workstation card with 16GB VRAM and lower power draw (~140W), ideal for tight builds.
I hope this helps!
If my response resolves your issue, please click “Accepted Solution” to help others find the answer. Also, don’t forget to click the “Yes” button to say thanks!
Take care and have a great day.
Max3Aj
HP Support