[GH-ISSUE #1263] Running Ollama for Orca2:13b on Limited VRAM: Model Configuration and Quantization Inquiry #645

Closed
opened 2026-04-12 10:20:23 -05:00 by GiteaMirror · 2 comments
Owner

Originally created by @derekhsu on GitHub (Nov 24, 2023).
Original GitHub issue: https://github.com/ollama/ollama/issues/1263

I successfully ran Ollama for Orca2:13b on my local machine, which has only 16GB of VRAM. However, I encountered difficulties running the Orca-2-13b model downloaded from Hugging Face due to insufficient VRAM. What model configuration does Ollama use to run Orca2:13b? Also, is it quantized?"

Originally created by @derekhsu on GitHub (Nov 24, 2023). Original GitHub issue: https://github.com/ollama/ollama/issues/1263 I successfully ran Ollama for Orca2:13b on my local machine, which has only 16GB of VRAM. However, I encountered difficulties running the Orca-2-13b model downloaded from Hugging Face due to insufficient VRAM. What model configuration does Ollama use to run Orca2:13b? Also, is it quantized?"
Author
Owner

@BruceMacD commented on GitHub (Nov 24, 2023):

HI @derekhsu the default Orca2:13b model is a 4-bit quantized model.

<!-- gh-comment-id:1826106954 --> @BruceMacD commented on GitHub (Nov 24, 2023): HI @derekhsu the default Orca2:13b model is a 4-bit quantized model.
Author
Owner

@mchiang0610 commented on GitHub (Mar 12, 2024):

Hi Derek, hope all is well! Can I assume your question is answered? Feel free to ask more! We're here to help.

Thanks!

<!-- gh-comment-id:1992509374 --> @mchiang0610 commented on GitHub (Mar 12, 2024): Hi Derek, hope all is well! Can I assume your question is answered? Feel free to ask more! We're here to help. Thanks!
Sign in to join this conversation.
1 Participants
Notifications
Due Date
No due date set.
Dependencies

No dependencies set.

Reference: github-starred/ollama#645