Skip to content

Conversation

@stevemason
Copy link
Contributor

I had an issue when using ollama targets where I could only use models which had been manually added to the hardcoded list of supported models (_OPTION_MAP). Example error output below, where the unknown key is listed as a valid key:

Error: Unknown Ollama key 'llama3.1:8b'. Valid keys: gemma3:1b, gemma3:4b, tinyllama:1.1b, nomic-embed-text:v1.5, llama3.1:8b

Given that ollama is queried for which models are available, I think it makes sense to use this query as the basis for which models are available for ollama targets.

I've modified the code to reflect this, and also removed references to _OPTION_MAP in the ollama_api module.

I also saw that when doing 'spikee list targets' it listed the first model as the default, so I've fixed the bahaviour to reflect this for ollama targets.

I've kept _DEFAULT_KEY for now to make this change easier, though I think it could probably be removed in the future.

Also, adding .sonar folder to .gitignore is helpful for those using sonarqube

@kyuz0
Copy link
Collaborator

kyuz0 commented Jan 5, 2026

Thank you for contributing this.

@kyuz0 kyuz0 merged commit 0ada1d6 into ReversecLabs:main Jan 5, 2026
1 check passed
@stevemason stevemason deleted the fix-ollama-model-selection branch January 6, 2026 11:46
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

2 participants