The processor module using ollama.
CURRENTLY THIS PLUGIN IS EXPERIMENTAL.
This plugin needs:
This module has some options.
-
endpoint
Ollama endpoint. Default:
http://localhost:11434
-
model
Model name. Specified model is needed to be installed. Default:
codellama
-
systemPrompt
(optional) System prompt for ollama. This prompt sets the context or behavior for the chat model. For example, you can use it to specify the role or instructions that the model should follow when processing the input.
let g:tataku_recipes = #{
\ sample: #{
\ processor: #{
\ name: 'ollama',
\ options: #{
\ endpoint: 'http://localhost:11434',
\ model: 'codellama',
\ },
\ }
\ }
\ }