Packages

liter_llm

1.8.2
1.11.2 1.11.1 1.11.0 1.10.1 1.10.0 1.9.3 1.9.0 1.9.0-rc.2 1.9.0-rc.1 1.8.2 1.8.1 1.8.0 1.7.6 1.7.5 1.7.4 1.7.3 1.7.2 1.7.1 1.6.4 1.6.3 1.6.2 1.5.1 1.5.0 1.4.1 1.4.0-rc.61 1.4.0-rc.60 1.4.0-rc.59 1.4.0-rc.58 1.4.0-rc.57 1.4.0-rc.56 1.4.0-rc.55 1.4.0-rc.53 1.4.0-rc.52 1.4.0-rc.50 1.4.0-rc.48 1.4.0-rc.47 1.4.0-rc.46 1.4.0-rc.45 1.4.0-rc.44 1.4.0-rc.43 1.4.0-rc.39 1.4.0-rc.38 1.4.0-rc.37 1.4.0-rc.36 1.4.0-rc.35 1.4.0-rc.34 1.4.0-rc.33 1.4.0-rc.32 1.4.0-rc.31 1.4.0-rc.30 1.4.0-rc.27 1.4.0-rc.26 1.4.0-rc.25 1.4.0-rc.24 1.4.0-rc.22 1.4.0-rc.21 1.4.0-rc.20 1.4.0-rc.19 1.4.0-rc.18 1.4.0-rc.17 1.4.0-rc.16 1.4.0-rc.14 1.4.0-rc.13 1.4.0-rc.11 1.2.2 1.2.1 1.2.0 1.1.1 1.1.0 1.0.0 1.0.0-rc.9 1.0.0-rc.8 1.0.0-rc.7 1.0.0-rc.6 1.0.0-rc.5 1.0.0-rc.4 1.0.0-rc.3 1.0.0-rc.2

Universal LLM API client with Rust-powered polyglot bindings.

Current section

Files

Jump to
Readme
78 Versions
3 Dependencies
Dependants
Files
Activity
liter_llm lib liter_llm response_format.ex
Raw

lib/liter_llm/response_format.ex 

# This file is auto-generated by alef — DO NOT EDIT.

# alef:hash:b159ec7bf40f2249b9f5696d5819688bca2310a2b26cc9acae8194e6255f8b03

# To regenerate: alef generate

# To verify freshness: alef verify --exit-code

defmodule LiterLlm.ResponseFormat do

  @moduledoc """

  Wire format for the chat completions `response_format` field.



  # Provider mapping



  - **OpenAI** (and OpenAI-compatible providers): emitted verbatim as

    `{"type": "json_schema", "json_schema": {...}}` per the

    chat-completions spec.

  - **Gemini / Vertex AI**: translated to

    `generationConfig.responseMimeType = "application/json"` and

    `generationConfig.responseSchema = <schema>`. The `name`,

    `description`, and `strict` fields are dropped — Gemini's

    structured-output API does not consume them.

  - **Anthropic**: no native JSON mode. A system instruction is

    prepended asking the model to respond with valid JSON.

    `strict` is advisory only; callers should still validate the

    returned JSON if the schema is load-bearing.



  # Example



  ```no_run

  # use liter_llm::types::{ResponseFormat, ChatCompletionRequest};

  # use serde_json::json;

  let request = ChatCompletionRequest {

      model: "gpt-4o".into(),

      messages: vec![],

      response_format: Some(ResponseFormat::json_schema(

          "PersonSchema",

          json!({ "type": "object", "properties": { "name": { "type": "string" } } }),

      )),

      ..Default::default()

  };

  # let _ = request;

  ```

  """



  @typedoc "Wire format for the chat completions `response_format` field."

  @type t :: term()



  @typedoc "Plain text output (default)."

  @type text :: :text

  @typedoc "Output must be valid JSON object (no schema validation)."

  @type json_object :: :json_object

  @typedoc "Output must conform to the specified JSON schema."

  @type json_schema :: %{type: :json_schema, json_schema: LiterLlm.JsonSchemaFormat.t()}

end