instructor-clj.core — org.clojars.kapil/instructor-clj 1.0.0-alpha

create-chat-completion clj

(create-chat-completion client-params)

Creates a chat completion using litellm-clj (supports multiple LLM providers).

Argument is a map with keys :messages, :model, :response-model, :provider, and optionally :api-key. :messages should be a vector of maps, each map representing a message with keys :role and :content. :model specifies the model to use (e.g., "gpt-3.5-turbo", "claude-3-opus-20240229", "gemini-pro"). :response-model is a Malli schema specifying the expected response structure. :provider (required) - must be specified explicitly (e.g., :openai, :anthropic, :gemini, :mistral, :ollama) :api-key (optional) - if not provided, will use environment variable OPENAI_API_KEY

Note: API keys can be set via environment variables:

• OPENAI_API_KEY for OpenAI models
• ANTHROPIC_API_KEY for Anthropic models
• GEMINI_API_KEY for Google Gemini models
• OPENROUTER_API_KEY for OpenRouter models

Example: (require '[instructor-clj.core :as ic])

(def User [:map [:name :string] [:age :int]])

(ic/create-chat-completion {:messages [{:role "user", :content "Jason Liu is 30 years old"}] :model "gpt-3.5-turbo" :provider :openai :response-model User})

Returns a map with extracted information in a structured format.

default-client-params clj

instruct clj

(instruct prompt
          response-schema
          &
          {:keys [api-key _max-tokens _model _temperature max-retries]
           :as client-params
           :or {max-retries 0}})

Attempts to obtain a valid response from the LLM based on the given prompt and schema, retrying up to max-retries times if necessary.

Note: API keys can be provided via :api-key parameter or OPENAI_API_KEY environment variable.

llm->response clj

(llm->response {:keys [prompt response-schema max-tokens model temperature
                       api-key provider]})

The function performs the LLM call and tries to destructure and get the actual response. Returns nil in cases where the LLM is not able to generate the expected response.

Supports multiple LLM providers through litellm-clj 0.3.0-alpha. Provider must be specified explicitly via :provider key (e.g., :openai, :anthropic, :gemini).

normalize-message clj

(normalize-message message)

Normalizes a message map to ensure :role is a keyword. litellm-clj 0.3.0-alpha requires roles to be keywords.

normalize-messages clj

(normalize-messages messages)

Normalizes a collection of messages to ensure all roles are keywords.

parse-generated-body clj

(parse-generated-body body)

Parses the body of a response generated by an LLM API call. Extracts and converts the message content into a Clojure map.

schema->system-prompt clj

(schema->system-prompt schema)

Converts a malli schema into JSON schema and generates a system prompt for responses