After checking with Local LLMs from Ollama always get the same error

[frankfullstack]

I tried to use Hashbrown with Angular but when I integrate with a local LLM I receive the following message when I try with llama3.1:8b model:

error:ResponseError: registry.ollama.ai/library/llama3.1:8b does not support thinking

or when I try with the following model: deepseek-r1:1.5b

error:ResponseError: think value "high" is not supported for this model

Seems like there are some issues when you try to use local Ollama models.

[MikeRyanDev]

We've got this fixed, but it's in a pre-release right now. Give 0.4.0-rc.1 a spin.

[frankfullstack]

Thanks so much @MikeRyanDev! You rock! 🚀