ollama - 💡(How to fix) Fix Ollama can't run hf.co/unsloth/Qwen3.5-35B-A3B-GGUF:Q8_0 [1 participants]

Even on unsloth webstie (https://unsloth.ai/docs/zh/mo-xing/qwen3.5), it says ollama is supported and provides the command line to run with ollama, I still got below error on Mac: Error: 500 Internal Server Error: unable to load model: /Users/elton/.ollama/models/blobs/sha256-3808866c016ab02b4adb26b873f7008a2cdd2c0704a39704050119ab0631db46

Fix Plan

The issue seems to be related to loading a large model with Ollama on a Mac. To fix this, we can try the following steps:

• Increase the memory allocation for the Ollama container
• Verify the model download and loading process

Here are the concrete steps:

1. Stop the current Ollama process:

pkill -f ollama

2. **Increase the memory allocation**:
   You can do this by setting the `--memory` flag when running Ollama. For example:
   ```bash
ollama run --memory 16G hf.co/unsloth/Qwen3.5-35B-A3B-GGUF:Q8_0

Replace 16G with the amount of memory you want to allocate. 3. Verify the model download and loading process: You can try downloading the model manually using the following command:

ollama pull hf.co/unsloth/Qwen3.5-35B-A3B-GGUF:Q8_0

Then, try running Ollama again with the --no-pull flag:

ollama run --no-pull hf.co/unsloth/Qwen3.5-35B-A3B-GGUF:Q8_0

Verification

To verify that the fix worked, you can check the Ollama logs for any error messages. If the model loads successfully, you should see a message indicating that the model is ready to use.

Extra Tips

• Make sure you have enough disk space and memory available on your Mac to load the model.
• If you continue to encounter issues, try updating Ollama to the latest version or seeking help from the Ollama community.