Liking cljdoc? Tell your friends :D
Clojure only.
ragtacts.llm.base
askcljmultimethod
Returns the answer to the question you asked the LLM.
Args:
-
q: The string or messages seq question you want to ask the LLM. If it were a sequence of messages, the items would have the following keys and string values:
:system,:user,:ai.- [{:system "You are a helpful assistant."}, {:user "Hello!"}]
-
params: A map of parameters to pass to the LLM.
:type: The type of LLM to use. Defaults to :open-ai.max-tokens: The maximum number of tokens that can be generated in the chat completion.temperature: What sampling temperature to use, between 0 and 2.top-p: An alternative to sampling with temperature, called nucleus sampling, where the model considers the results of the tokens with top-p probability mass.:tools: List of function Vars to use as tools for the LLM.:as: If you use:valuesin the:asoption when tool is invoked with the tools option, it will return the resulting list of following map of tool calls:- key: function name
- value: result of the function call
Returns:
- String: The answer to the question you asked the LLM.
Example:
(ask "Hello!")
(ask "Hello!" {:type :open-ai})
(ask "Hello!" {:type :open-ai :model "gpt-4o"})
(ask [{:system "You are a helpful assistant."}, {:user "Hello!"}])
;; tools
(defn ^{:desc "Get the current weather in a given location"} get-current-weather
[^{:type "string" :desc "The city and state, e.g. San Francisco, CA"} location]
...))
(ask "What's the weather like in San Francisco, Tokyo, and Paris?"
{:tools [#'get-current-weather]})
Returns the answer to the question you asked the LLM.
Args:
- q: The string or messages seq question you want to ask the LLM. If it were a sequence of
messages, the items would have the following keys and string values: `:system`, `:user`, `:ai`.
- [{:system "You are a helpful assistant."}, {:user "Hello!"}]
- params: A map of parameters to pass to the LLM.
- `:type`: The type of LLM to use. Defaults to :open-ai.
- `max-tokens`: The maximum number of tokens that can be generated in the chat completion.
- `temperature`: What sampling temperature to use, between 0 and 2.
- `top-p`: An alternative to sampling with temperature, called nucleus sampling,
where the model considers the results of the tokens with top-p probability mass.
- `:tools`: List of function Vars to use as tools for the LLM.
- `:as`: If you use `:values` in the `:as` option when tool is invoked with the tools
option, it will return the resulting list of following map of tool calls:
- key: function name
- value: result of the function call
Returns:
- String: The answer to the question you asked the LLM.
Example:
```clojure
(ask "Hello!")
(ask "Hello!" {:type :open-ai})
(ask "Hello!" {:type :open-ai :model "gpt-4o"})
(ask [{:system "You are a helpful assistant."}, {:user "Hello!"}])
;; tools
(defn ^{:desc "Get the current weather in a given location"} get-current-weather
[^{:type "string" :desc "The city and state, e.g. San Francisco, CA"} location]
...))
(ask "What's the weather like in San Francisco, Tokyo, and Paris?"
{:tools [#'get-current-weather]})
```
cljdoc is a website building & hosting documentation for Clojure/Script libraries
× close