How to run a local LLM on the NVIDIA RTX PRO 6000 Blackwell
A step-by-step setup for beginners. With 96 GB of graphics memory, the best everyday model for this card is gpt-oss 120B, which runs very fast: answers appear almost instantly.
What to run on the RTX PRO 6000 Blackwell
The best model for each job that fits fully in memory at the default Q4_K_M download and runs at a comfortable reading speed.
| Job | Model | Speed | Ollama command |
|---|---|---|---|
| Chat and everyday questions |  gpt-oss 120B | Very fast: answers appear almost instantly | ollama run gpt-oss:120b |
| Help with code |  Qwen3 Coder Next 80B-A3B | Very fast: answers appear almost instantly | ollama run qwen3-coder-next |
| Writing and editing |  Llama 3.3 70B | Faster than you can read | ollama run llama3.3:70b |
| Questions about my documents |  Gemma 4 26B-A4B | Very fast: answers appear almost instantly | ollama run gemma4:26b |
| Maths, logic and homework | gpt-oss 120B | Very fast: answers appear almost instantly | ollama run gpt-oss:120b |
Set up gpt-oss 120B in about ten minutes
Pick one app. LM Studio is easiest if you've never used a terminal.
LM Studio on Windows
- Go to lmstudio.ai and click Download for Windows.
- Open the downloaded installer and click through it. LM Studio opens when it finishes.
- If it offers to download a first model, you can skip that. Click the search (magnifying glass) icon in the left sidebar to open Discover.
- Type gpt-oss 120B in the search box and pick the result from a well-known publisher such as lmstudio-community.
- Choose the Q4_K_M download if a list of versions appears. LM Studio marks versions that fit your GPU with a green "Full GPU offload possible" label. Click Download and wait for it to finish.
- Open the Chat tab, click Select a model to load at the top, pick gpt-oss 120B, and type your first message.
Ollama on Windows
- Go to ollama.com/download and download the Windows installer.
- Run OllamaSetup.exe. When it finishes, Ollama keeps running in the background (look for the llama icon by the clock).
- Open PowerShell (search for "PowerShell" in the Start menu) and run the command below. The first time, it downloads gpt-oss 120B, which can take a while.
ollama run gpt-oss:120b- When you see
>>>, type a message and press Enter. Type/byeto quit. Ollama's own app window can also chat with any model you have downloaded.
On Linux? Follow the LM Studio or Ollama Linux guide instead.
If something goes wrong
The four problems beginners hit most.
Out of memory or won't load
The model plus its working memory doesn't fit in your graphics memory. Here is how to make it fit, from the easiest fix to the most advanced.
It's really slow
Most slowness comes from the model not fitting fully in graphics memory, or from "thinking" models reasoning before they answer.
It's not using my GPU
If your graphics card sits idle while the CPU does all the work, one of these is usually the reason.
Weird, wrong or repeating answers
Local models can be excellent, but small or badly configured ones make mistakes that paid assistants don't. Here is what to check.
Run it in the cloud
Want a bigger model than the RTX PRO 6000 Blackwell can hold?
Rent a GPU with more memory by the hour, or see whether upgrading or renting is cheaper for you.
Frequently asked questions
What is the best local LLM for the NVIDIA RTX PRO 6000 Blackwell?
For everyday chat, gpt-oss 120B: it fits in the NVIDIA RTX PRO 6000 Blackwell's 96 GB at Q4_K_M and runs at about 348 tokens per second.
Should I use LM Studio or Ollama on the NVIDIA RTX PRO 6000 Blackwell?
Either works. LM Studio is a point-and-click app and the easiest start; Ollama is a one-line command that other apps and coding tools can connect to.
Can the NVIDIA RTX PRO 6000 Blackwell run a 70B model?
Yes, at Q8_0, at about 14 tokens per second.