LLM not working getting error

I tried both the llama3.2:1b and llama3.2:3b models but i am getting error like this when i am trying to run “ollama run llama3.2:3b” and “ollama run llama3.2:1b”. This is the first time i am facing this issue, before i didn’t get such errors. I am using NVIDIA Jetson orin Nano Developer kit 8GB. Can someone someone please help me with this.

*** Please note that this reply is generated by LLM automatically ***

Based on the provided information, it appears that you are experiencing issues with running the LLaMA models (llama321b and llama323b) on your NVIDIA Jetson Orin Nano Developer Kit 8GB. The error message you provided is not explicitly stated, but I’ll try to offer some general guidance and potential solutions.

Firstly, ensure that you have the correct version of the LLaMA model and the necessary dependencies installed on your Jetson Orin Nano Developer Kit. You can check the version of the LLaMA model by running the command ollama --version.

Since you mentioned that you didn’t encounter this issue before, it’s possible that there might be a compatibility issue with the latest version of the LLaMA model or the JetPack version on your Jetson Orin Nano Developer Kit.

Here are a few potential solutions you can try:

  1. Check the JetPack version: Ensure that you are running the latest version of JetPack on your Jetson Orin Nano Developer Kit. You can check the version by running the command dpkg -l | grep jetpack. If you are not running the latest version, update JetPack to the latest version.
  2. Verify the LLaMA model version: Check the version of the LLaMA model you are trying to run. You can do this by running the command ollama --version. If you are not running the latest version, update the LLaMA model to the latest version.
  3. Check the dependencies: Ensure that you have the necessary dependencies installed on your Jetson Orin Nano Developer Kit. You can check the dependencies by running the command dpkg -l | grep dependency_name.
  4. Try a different model: If you are still experiencing issues with the LLaMA models, try running a different model to see if the issue is specific to the LLaMA models.
  5. Check the system logs: Check the system logs to see if there are any error messages that might indicate the cause of the issue. You can check the system logs by running the command dmesg | grep error.

In terms of comparing the Jetson Orin Nano Developer Kit 8GB with other Jetson modules, here are some key differences:

  • Jetson Orin Nano Series: The Jetson Orin Nano Series includes the Jetson Orin Nano 8GB (P3767-0003), Jetson Orin Nano 4GB (P3767-0004), and Jetson Orin Nano 8GB with SD card slot (P3767-0005). The Jetson Orin Nano Developer Kit 8GB (P3766) includes the P3767-0005 module.
  • Jetson Orin NX Series: The Jetson Orin NX Series includes the Jetson Orin NX 16GB (P3767-0000) and Jetson Orin NX 8GB (P3767-0001). The Jetson Orin NX Series is not compatible with the Jetson Orin Nano Developer Kit 8GB.
  • Jetson AGX Orin Series: The Jetson AGX Orin Series includes the Jetson AGX Orin 32GB (P3701-0000), Jetson AGX Orin 32GB (P3701-0004), and Jetson AGX Orin 64GB (P3701-0005). The Jetson AGX Orin Series is not compatible with the Jetson Orin Nano Developer Kit 8GB.

In summary, the issue you are experiencing with running the LLaMA models on your Jetson Orin Nano Developer Kit 8GB might be related to compatibility issues with the latest version of the LLaMA model or the JetPack version on your device. Try updating JetPack and the LLaMA model to the latest versions, and check the dependencies and system logs for any error messages.

*** Please give the thumbs up if you get answers you like. Or provide feedback to help us improve the answer. ***

Hi,

Could you check which BSP you are using?

$ cat /etc/nv_tegra_release

There is a known issue in r36.4.7.
If you are using the BSP, please upgrade your device to JetPack 6.2.2/r36.5.

Thanks.