ollama - 💡(How to fix) Fix Ollama tries to pull when it shouldn't [2 comments, 2 participants]

Error: OSS setup failed: Pull failed: pull model manifest: file does not exist

• Launching codex -p myprofile also fails, even if I have configured base_url = "http://localhost:11434/v1". After the "model metadata not found" error I get:

{"error":{"message":"registry.ollama.ai/library/mymodel:latest does not support tools","type":"api_error","param":null,"code":null}}

Fix Plan

The fix involves modifying the ollama launch codex command to use the local model instead of pulling from the cloud.

• Set the OLLAMA_MODEL_REGISTRY environment variable to point to the local model registry.
• Use the --config flag with ollama launch codex to specify the local model configuration.

Example:

OLLAMA_MODEL_REGISTRY=file:///path/to/local/model/registry ollama launch codex --config /path/to/model/config

Alternatively, you can also modify the ollama configuration file to use the local model registry by default.

• Add the following lines to your ~/.ollama/config file:

model_registry: file:///path/to/local/model/registry

Then, run the ollama launch codex command without any additional flags.

Verification

To verify that the fix worked, run the ollama launch codex command and check that it uses the local model instead of pulling from the cloud. You can also check the ollama logs to ensure that the local model registry is being used.

Extra Tips

• Make sure to update the OLLAMA_MODEL_REGISTRY environment variable or the ~/.ollama/config file to point to the correct location of your local model registry.
• If you are using a custom ollama configuration file, make sure to update the model_registry setting accordingly.