# AI

## Execute AI model

**post** `/accounts/{account_id}/ai/run/{model_name}`

This endpoint provides users with the capability to run specific AI models on-demand.

By submitting the required input data, users can receive real-time predictions or results generated by the chosen AI
model. The endpoint supports various AI model types, ensuring flexibility and adaptability for diverse use cases.

Model specific inputs available in [Cloudflare Docs](https://developers.cloudflare.com/workers-ai/models/).

### Path Parameters

- `account_id: string`

- `model_name: string`

### Body Parameters

- `body: optional object { text }  or object { prompt, guidance, height, 8 more }  or object { prompt, lang }  or 12 more`

  - `TextClassification = object { text }`

    - `text: string`

      The text that you want to classify

  - `TextToImage = object { prompt, guidance, height, 8 more }`

    - `prompt: string`

      A text description of the image you want to generate

    - `guidance: optional number`

      Controls how closely the generated image should adhere to the prompt; higher values make the image more aligned with the prompt

    - `height: optional number`

      The height of the generated image in pixels

    - `image: optional array of number`

      For use with img2img tasks. An array of integers that represent the image data constrained to 8-bit unsigned integer values

    - `image_b64: optional string`

      For use with img2img tasks. A base64-encoded string of the input image

    - `mask: optional array of number`

      An array representing An array of integers that represent mask image data for inpainting constrained to 8-bit unsigned integer values

    - `negative_prompt: optional string`

      Text describing elements to avoid in the generated image

    - `num_steps: optional number`

      The number of diffusion steps; higher values can improve quality but take longer

    - `seed: optional number`

      Random seed for reproducibility of the image generation

    - `strength: optional number`

      A value between 0 and 1 indicating how strongly to apply the transformation during img2img tasks; lower values make the output closer to the input image

    - `width: optional number`

      The width of the generated image in pixels

  - `TextToSpeech = object { prompt, lang }`

    - `prompt: string`

      A text description of the audio you want to generate

    - `lang: optional string`

      The speech language (e.g., 'en' for English, 'fr' for French). Defaults to 'en' if not specified

  - `TextEmbeddings = object { text }`

    - `text: string or array of string`

      The text to embed

      - `string`

        The text to embed

      - `array of string`

        Batch of text values to embed

  - `AutomaticSpeechRecognition = object { audio, source_lang, target_lang }`

    - `audio: array of number`

      An array of integers that represent the audio data constrained to 8-bit unsigned integer values

    - `source_lang: optional string`

      The language of the recorded audio

    - `target_lang: optional string`

      The language to translate the transcription into. Currently only English is supported.

  - `ImageClassification = object { image }`

    - `image: array of number`

      An array of integers that represent the image data constrained to 8-bit unsigned integer values

  - `ObjectDetection = object { image }`

    - `image: optional array of number`

      An array of integers that represent the image data constrained to 8-bit unsigned integer values

  - `Prompt = object { prompt, frequency_penalty, lora, 10 more }`

    - `prompt: string`

      The input text prompt for the model to generate a response.

    - `frequency_penalty: optional number`

      Decreases the likelihood of the model repeating the same lines verbatim.

    - `lora: optional string`

      Name of the LoRA (Low-Rank Adaptation) model to fine-tune the base model.

    - `max_tokens: optional number`

      The maximum number of tokens to generate in the response.

    - `presence_penalty: optional number`

      Increases the likelihood of the model introducing new topics.

    - `raw: optional boolean`

      If true, a chat template is not applied and you must adhere to the specific model's expected formatting.

    - `repetition_penalty: optional number`

      Penalty for repeated tokens; higher values discourage repetition.

    - `response_format: optional object { json_schema, type }`

      - `json_schema: optional unknown`

      - `type: optional "json_object" or "json_schema"`

        - `"json_object"`

        - `"json_schema"`

    - `seed: optional number`

      Random seed for reproducibility of the generation.

    - `stream: optional boolean`

      If true, the response will be streamed back incrementally using SSE, Server Sent Events.

    - `temperature: optional number`

      Controls the randomness of the output; higher values produce more random results.

    - `top_k: optional number`

      Limits the AI to choose from the top 'k' most probable words. Lower values make responses more focused; higher values introduce more variety and potential surprises.

    - `top_p: optional number`

      Adjusts the creativity of the AI's responses by controlling how many possible words it considers. Lower values make outputs more predictable; higher values allow for more varied and creative responses.

  - `TextGeneration = object { messages, frequency_penalty, functions, 11 more }`

    - `messages: array of object { content, role }`

      An array of message objects representing the conversation history.

      - `content: string or array of object { text, type }`

        The content of the message as a string.

        - `string`

          The content of the message as a string.

        - `array of object { text, type }`

          Array of text content parts.

          - `text: optional string`

            Text content

          - `type: optional string`

            Type of the content (text)

      - `role: string`

        The role of the message sender (e.g., 'user', 'assistant', 'system', 'tool').

    - `frequency_penalty: optional number`

      Decreases the likelihood of the model repeating the same lines verbatim.

    - `functions: optional array of object { code, name }`

      - `code: string`

      - `name: string`

    - `max_tokens: optional number`

      The maximum number of tokens to generate in the response.

    - `presence_penalty: optional number`

      Increases the likelihood of the model introducing new topics.

    - `raw: optional boolean`

      If true, a chat template is not applied and you must adhere to the specific model's expected formatting.

    - `repetition_penalty: optional number`

      Penalty for repeated tokens; higher values discourage repetition.

    - `response_format: optional object { json_schema, type }`

      - `json_schema: optional unknown`

      - `type: optional "json_object" or "json_schema"`

        - `"json_object"`

        - `"json_schema"`

    - `seed: optional number`

      Random seed for reproducibility of the generation.

    - `stream: optional boolean`

      If true, the response will be streamed back incrementally using SSE, Server Sent Events.

    - `temperature: optional number`

      Controls the randomness of the output; higher values produce more random results.

    - `tools: optional array of object { description, name, parameters }  or object { function, type }`

      A list of tools available for the assistant to use.

      - `object { description, name, parameters }`

        - `description: string`

          A brief description of what the tool does.

        - `name: string`

          The name of the tool. More descriptive the better.

        - `parameters: object { properties, type, required }`

          Schema defining the parameters accepted by the tool.

          - `properties: map[object { description, type } ]`

            Definitions of each parameter.

            - `description: string`

              A description of the expected parameter.

            - `type: string`

              The data type of the parameter.

          - `type: string`

            The type of the parameters object (usually 'object').

          - `required: optional array of string`

            List of required parameter names.

      - `Function = object { function, type }`

        - `function: object { description, name, parameters }`

          Details of the function tool.

          - `description: string`

            A brief description of what the function does.

          - `name: string`

            The name of the function.

          - `parameters: object { properties, type, required }`

            Schema defining the parameters accepted by the function.

            - `properties: map[object { description, type } ]`

              Definitions of each parameter.

              - `description: string`

                A description of the expected parameter.

              - `type: string`

                The data type of the parameter.

            - `type: string`

              The type of the parameters object (usually 'object').

            - `required: optional array of string`

              List of required parameter names.

        - `type: string`

          Specifies the type of tool (e.g., 'function').

    - `top_k: optional number`

      Limits the AI to choose from the top 'k' most probable words. Lower values make responses more focused; higher values introduce more variety and potential surprises.

    - `top_p: optional number`

      Adjusts the creativity of the AI's responses by controlling how many possible words it considers. Lower values make outputs more predictable; higher values allow for more varied and creative responses.

  - `Translation = object { target_lang, text, source_lang }`

    - `target_lang: string`

      The language code to translate the text into (e.g., 'es' for Spanish)

    - `text: string`

      The text to be translated

    - `source_lang: optional string`

      The language code of the source text (e.g., 'en' for English). Defaults to 'en' if not specified

  - `Summarization = object { input_text, max_length }`

    - `input_text: string`

      The text that you want the model to summarize

    - `max_length: optional number`

      The maximum length of the generated summary in tokens

  - `ImageToText = object { image, frequency_penalty, max_tokens, 8 more }`

    - `image: array of number`

      An array of integers that represent the image data constrained to 8-bit unsigned integer values

    - `frequency_penalty: optional number`

      Decreases the likelihood of the model repeating the same lines verbatim.

    - `max_tokens: optional number`

      The maximum number of tokens to generate in the response.

    - `presence_penalty: optional number`

      Increases the likelihood of the model introducing new topics.

    - `prompt: optional string`

      The input text prompt for the model to generate a response.

    - `raw: optional boolean`

      If true, a chat template is not applied and you must adhere to the specific model's expected formatting.

    - `repetition_penalty: optional number`

      Penalty for repeated tokens; higher values discourage repetition.

    - `seed: optional number`

      Random seed for reproducibility of the generation.

    - `temperature: optional number`

      Controls the randomness of the output; higher values produce more random results.

    - `top_k: optional number`

      Limits the AI to choose from the top 'k' most probable words. Lower values make responses more focused; higher values introduce more variety and potential surprises.

    - `top_p: optional number`

      Controls the creativity of the AI's responses by adjusting how many possible words it considers. Lower values make outputs more predictable; higher values allow for more varied and creative responses.

  - `object { image, prompt, frequency_penalty, 8 more }`

    - `image: string`

      Image in base64 encoded format.

    - `prompt: string`

      The input text prompt for the model to generate a response.

    - `frequency_penalty: optional number`

      Decreases the likelihood of the model repeating the same lines verbatim.

    - `ignore_eos: optional boolean`

      Whether to ignore the EOS token and continue generating tokens after the EOS token is generated.

    - `max_tokens: optional number`

      The maximum number of tokens to generate in the response.

    - `presence_penalty: optional number`

      Increases the likelihood of the model introducing new topics.

    - `repetition_penalty: optional number`

      Penalty for repeated tokens; higher values discourage repetition.

    - `seed: optional number`

      Random seed for reproducibility of the generation.

    - `temperature: optional number`

      Controls the randomness of the output; higher values produce more random results.

    - `top_k: optional number`

      Limits the AI to choose from the top 'k' most probable words. Lower values make responses more focused; higher values introduce more variety and potential surprises.

    - `top_p: optional number`

      Controls the creativity of the AI's responses by adjusting how many possible words it considers. Lower values make outputs more predictable; higher values allow for more varied and creative responses.

  - `ImageTextToText = object { image, messages, frequency_penalty, 8 more }`

    - `image: string`

      Image in base64 encoded format.

    - `messages: array of object { content, role }`

      An array of message objects representing the conversation history.

      - `content: string or array of object { type, image_url, text }`

        The content of the message as a string.

        - `string`

          The content of the message as a string.

        - `array of object { type, image_url, text }`

          Array of content parts (text, image_url, etc.).

          - `type: string`

            Type of the content part (e.g. 'text', 'image_url').

          - `image_url: optional object { url }`

            Image URL object (when type is 'image_url').

            - `url: string`

              Image URI with data (e.g. data:image/jpeg;base64,/9j/...).

          - `text: optional string`

            Text content (when type is 'text').

      - `role: string`

        The role of the message sender (e.g., 'user', 'assistant', 'system', 'tool').

    - `frequency_penalty: optional number`

      Decreases the likelihood of the model repeating the same lines verbatim.

    - `ignore_eos: optional boolean`

      Whether to ignore the EOS token and continue generating tokens after the EOS token is generated.

    - `max_tokens: optional number`

      The maximum number of tokens to generate in the response.

    - `presence_penalty: optional number`

      Increases the likelihood of the model introducing new topics.

    - `repetition_penalty: optional number`

      Penalty for repeated tokens; higher values discourage repetition.

    - `seed: optional number`

      Random seed for reproducibility of the generation.

    - `temperature: optional number`

      Controls the randomness of the output; higher values produce more random results.

    - `top_k: optional number`

      Limits the AI to choose from the top 'k' most probable words. Lower values make responses more focused; higher values introduce more variety and potential surprises.

    - `top_p: optional number`

      Controls the creativity of the AI's responses by adjusting how many possible words it considers. Lower values make outputs more predictable; higher values allow for more varied and creative responses.

  - `MultimodalEmbeddings = object { image, text }`

    - `image: optional string`

      Image in base64 encoded format.

    - `text: optional array of string`

### Returns

- `result: optional array of object { label, score }  or string or object { audio }  or 12 more`

  An array of classification results for the input text

  - `TextClassification = array of object { label, score }`

    An array of classification results for the input text

    - `label: optional string`

      The classification label assigned to the text (e.g., 'POSITIVE' or 'NEGATIVE')

    - `score: optional number`

      Confidence score indicating the likelihood that the text belongs to the specified label

  - `TextToImage = string`

    The generated image in PNG format

  - `Audio = object { audio }`

    - `audio: optional string`

      The generated audio in MP3 format, base64-encoded

  - `string`

    The generated audio in MP3 format

  - `TextEmbeddings = object { data, shape }`

    - `data: optional array of array of number`

      Embeddings of the requested text values

    - `shape: optional array of number`

  - `AutomaticSpeechRecognition = object { text, vtt, word_count, words }`

    - `text: string`

      The transcription

    - `vtt: optional string`

    - `word_count: optional number`

    - `words: optional array of object { end, start, word }`

      - `end: optional number`

        The ending second when the word completes

      - `start: optional number`

        The second this word begins in the recording

      - `word: optional string`

  - `ImageClassification = array of object { label, score }`

    - `label: optional string`

      The predicted category or class for the input image based on analysis

    - `score: optional number`

      A confidence value, between 0 and 1, indicating how certain the model is about the predicted label

  - `ObjectDetection = array of object { box, label, score }`

    An array of detected objects within the input image

    - `box: optional object { xmax, xmin, ymax, ymin }`

      Coordinates defining the bounding box around the detected object

      - `xmax: optional number`

        The x-coordinate of the bottom-right corner of the bounding box

      - `xmin: optional number`

        The x-coordinate of the top-left corner of the bounding box

      - `ymax: optional number`

        The y-coordinate of the bottom-right corner of the bounding box

      - `ymin: optional number`

        The y-coordinate of the top-left corner of the bounding box

    - `label: optional string`

      The class label or name of the detected object

    - `score: optional number`

      Confidence score indicating the likelihood that the detection is correct

  - `object { response, tool_calls, usage }`

    - `response: string`

      The generated text response from the model

    - `tool_calls: optional array of object { arguments, name }`

      An array of tool calls requests made during the response generation

      - `arguments: optional unknown`

        The arguments passed to be passed to the tool call request

      - `name: optional string`

        The name of the tool to be called

    - `usage: optional object { completion_tokens, prompt_tokens, total_tokens }`

      Usage statistics for the inference request

      - `completion_tokens: optional number`

        Total number of tokens in output

      - `prompt_tokens: optional number`

        Total number of tokens in input

      - `total_tokens: optional number`

        Total number of input and output tokens

  - `string`

  - `Translation = object { translated_text }`

    - `translated_text: optional string`

      The translated text in the target language

  - `Summarization = object { summary }`

    - `summary: optional string`

      The summarized version of the input text

  - `ImageToText = object { description }`

    - `description: optional string`

  - `ImageTextToText = object { description }`

    - `description: optional string`

  - `MultimodalEmbeddings = object { data, shape }`

    - `data: optional array of array of number`

    - `shape: optional array of number`

### Example

```http
curl https://api.cloudflare.com/client/v4/accounts/$ACCOUNT_ID/ai/run/$MODEL_NAME \
    -X POST \
    -H "Authorization: Bearer $CLOUDFLARE_API_TOKEN"
```

#### Response

```json
{
  "result": [
    {
      "label": "label",
      "score": 0
    }
  ]
}
```

## Domain Types

### AI Run Response

- `AIRunResponse = array of object { label, score }  or string or object { audio }  or 12 more`

  An array of classification results for the input text

  - `TextClassification = array of object { label, score }`

    An array of classification results for the input text

    - `label: optional string`

      The classification label assigned to the text (e.g., 'POSITIVE' or 'NEGATIVE')

    - `score: optional number`

      Confidence score indicating the likelihood that the text belongs to the specified label

  - `TextToImage = string`

    The generated image in PNG format

  - `Audio = object { audio }`

    - `audio: optional string`

      The generated audio in MP3 format, base64-encoded

  - `string`

    The generated audio in MP3 format

  - `TextEmbeddings = object { data, shape }`

    - `data: optional array of array of number`

      Embeddings of the requested text values

    - `shape: optional array of number`

  - `AutomaticSpeechRecognition = object { text, vtt, word_count, words }`

    - `text: string`

      The transcription

    - `vtt: optional string`

    - `word_count: optional number`

    - `words: optional array of object { end, start, word }`

      - `end: optional number`

        The ending second when the word completes

      - `start: optional number`

        The second this word begins in the recording

      - `word: optional string`

  - `ImageClassification = array of object { label, score }`

    - `label: optional string`

      The predicted category or class for the input image based on analysis

    - `score: optional number`

      A confidence value, between 0 and 1, indicating how certain the model is about the predicted label

  - `ObjectDetection = array of object { box, label, score }`

    An array of detected objects within the input image

    - `box: optional object { xmax, xmin, ymax, ymin }`

      Coordinates defining the bounding box around the detected object

      - `xmax: optional number`

        The x-coordinate of the bottom-right corner of the bounding box

      - `xmin: optional number`

        The x-coordinate of the top-left corner of the bounding box

      - `ymax: optional number`

        The y-coordinate of the bottom-right corner of the bounding box

      - `ymin: optional number`

        The y-coordinate of the top-left corner of the bounding box

    - `label: optional string`

      The class label or name of the detected object

    - `score: optional number`

      Confidence score indicating the likelihood that the detection is correct

  - `object { response, tool_calls, usage }`

    - `response: string`

      The generated text response from the model

    - `tool_calls: optional array of object { arguments, name }`

      An array of tool calls requests made during the response generation

      - `arguments: optional unknown`

        The arguments passed to be passed to the tool call request

      - `name: optional string`

        The name of the tool to be called

    - `usage: optional object { completion_tokens, prompt_tokens, total_tokens }`

      Usage statistics for the inference request

      - `completion_tokens: optional number`

        Total number of tokens in output

      - `prompt_tokens: optional number`

        Total number of tokens in input

      - `total_tokens: optional number`

        Total number of input and output tokens

  - `string`

  - `Translation = object { translated_text }`

    - `translated_text: optional string`

      The translated text in the target language

  - `Summarization = object { summary }`

    - `summary: optional string`

      The summarized version of the input text

  - `ImageToText = object { description }`

    - `description: optional string`

  - `ImageTextToText = object { description }`

    - `description: optional string`

  - `MultimodalEmbeddings = object { data, shape }`

    - `data: optional array of array of number`

    - `shape: optional array of number`

# Finetunes

## List Finetunes

**get** `/accounts/{account_id}/ai/finetunes`

Lists all fine-tuning jobs created by the account, including status and metrics.

### Path Parameters

- `account_id: string`

### Returns

- `result: object { id, created_at, model, 3 more }`

  - `id: string`

  - `created_at: string`

  - `model: string`

  - `modified_at: string`

  - `name: string`

  - `description: optional string`

- `success: boolean`

### Example

```http
curl https://api.cloudflare.com/client/v4/accounts/$ACCOUNT_ID/ai/finetunes \
    -H "Authorization: Bearer $CLOUDFLARE_API_TOKEN"
```

#### Response

```json
{
  "result": {
    "id": "182bd5e5-6e1a-4fe4-a799-aa6d9a6ab26e",
    "created_at": "2019-12-27T18:11:19.117Z",
    "model": "model",
    "modified_at": "2019-12-27T18:11:19.117Z",
    "name": "name",
    "description": "description"
  },
  "success": true
}
```

## Create a new Finetune

**post** `/accounts/{account_id}/ai/finetunes`

Creates a new fine-tuning job for a Workers AI model using custom training data.

### Path Parameters

- `account_id: string`

### Body Parameters

- `model: string`

- `name: string`

- `description: optional string`

- `public: optional boolean`

### Returns

- `result: object { id, created_at, model, 4 more }`

  - `id: string`

  - `created_at: string`

  - `model: string`

  - `modified_at: string`

  - `name: string`

  - `public: boolean`

  - `description: optional string`

- `success: boolean`

### Example

```http
curl https://api.cloudflare.com/client/v4/accounts/$ACCOUNT_ID/ai/finetunes \
    -H 'Content-Type: application/json' \
    -H "Authorization: Bearer $CLOUDFLARE_API_TOKEN" \
    -d '{
          "model": "model",
          "name": "name"
        }'
```

#### Response

```json
{
  "result": {
    "id": "182bd5e5-6e1a-4fe4-a799-aa6d9a6ab26e",
    "created_at": "2019-12-27T18:11:19.117Z",
    "model": "model",
    "modified_at": "2019-12-27T18:11:19.117Z",
    "name": "name",
    "public": true,
    "description": "description"
  },
  "success": true
}
```

## Domain Types

### Finetune List Response

- `FinetuneListResponse = object { id, created_at, model, 3 more }`

  - `id: string`

  - `created_at: string`

  - `model: string`

  - `modified_at: string`

  - `name: string`

  - `description: optional string`

### Finetune Create Response

- `FinetuneCreateResponse = object { id, created_at, model, 4 more }`

  - `id: string`

  - `created_at: string`

  - `model: string`

  - `modified_at: string`

  - `name: string`

  - `public: boolean`

  - `description: optional string`

# Assets

## Upload a Finetune Asset

**post** `/accounts/{account_id}/ai/finetunes/{finetune_id}/finetune-assets`

Uploads training data assets for a Workers AI fine-tuning job.

### Path Parameters

- `account_id: string`

- `finetune_id: string`

### Returns

- `success: boolean`

### Example

```http
curl https://api.cloudflare.com/client/v4/accounts/$ACCOUNT_ID/ai/finetunes/$FINETUNE_ID/finetune-assets \
    -X POST \
    -H "Authorization: Bearer $CLOUDFLARE_API_TOKEN"
```

#### Response

```json
{
  "success": true
}
```

## Domain Types

### Asset Create Response

- `AssetCreateResponse = object { success }`

  - `success: boolean`

# Public

## List Public Finetunes

**get** `/accounts/{account_id}/ai/finetunes/public`

Lists publicly available fine-tuned models that can be used with Workers AI.

### Path Parameters

- `account_id: string`

### Query Parameters

- `limit: optional number`

  Pagination Limit

- `offset: optional number`

  Pagination Offset

- `orderBy: optional string`

  Order By Column Name

### Returns

- `result: array of object { id, created_at, model, 4 more }`

  - `id: string`

  - `created_at: string`

  - `model: string`

  - `modified_at: string`

  - `name: string`

  - `public: boolean`

  - `description: optional string`

- `success: boolean`

### Example

```http
curl https://api.cloudflare.com/client/v4/accounts/$ACCOUNT_ID/ai/finetunes/public \
    -H "Authorization: Bearer $CLOUDFLARE_API_TOKEN"
```

#### Response

```json
{
  "result": [
    {
      "id": "182bd5e5-6e1a-4fe4-a799-aa6d9a6ab26e",
      "created_at": "2019-12-27T18:11:19.117Z",
      "model": "model",
      "modified_at": "2019-12-27T18:11:19.117Z",
      "name": "name",
      "public": true,
      "description": "description"
    }
  ],
  "success": true
}
```

## Domain Types

### Public List Response

- `PublicListResponse = object { id, created_at, model, 4 more }`

  - `id: string`

  - `created_at: string`

  - `model: string`

  - `modified_at: string`

  - `name: string`

  - `public: boolean`

  - `description: optional string`

# Authors

## Author Search

**get** `/accounts/{account_id}/ai/authors/search`

Searches Workers AI models by author or organization name.

### Path Parameters

- `account_id: string`

### Returns

- `errors: array of unknown`

- `messages: array of string`

- `result: array of unknown`

- `success: boolean`

### Example

```http
curl https://api.cloudflare.com/client/v4/accounts/$ACCOUNT_ID/ai/authors/search \
    -H "Authorization: Bearer $CLOUDFLARE_API_TOKEN"
```

#### Response

```json
{
  "errors": [
    {}
  ],
  "messages": [
    "string"
  ],
  "result": [
    {}
  ],
  "success": true
}
```

## Domain Types

### Author List Response

- `AuthorListResponse = unknown`

# Tasks

## Task Search

**get** `/accounts/{account_id}/ai/tasks/search`

Searches Workers AI models by task type (e.g., text-generation, embeddings).

### Path Parameters

- `account_id: string`

### Returns

- `errors: array of unknown`

- `messages: array of string`

- `result: array of unknown`

- `success: boolean`

### Example

```http
curl https://api.cloudflare.com/client/v4/accounts/$ACCOUNT_ID/ai/tasks/search \
    -H "Authorization: Bearer $CLOUDFLARE_API_TOKEN"
```

#### Response

```json
{
  "errors": [
    {}
  ],
  "messages": [
    "string"
  ],
  "result": [
    {}
  ],
  "success": true
}
```

## Domain Types

### Task List Response

- `TaskListResponse = unknown`

# Models

## Model Search

**get** `/accounts/{account_id}/ai/models/search`

Searches Workers AI models by name or description.

### Path Parameters

- `account_id: string`

### Query Parameters

- `author: optional string`

  Filter by Author

- `hide_experimental: optional boolean`

  Filter to hide experimental models

- `page: optional number`

- `per_page: optional number`

- `search: optional string`

  Search

- `source: optional number`

  Filter by Source Id

- `task: optional string`

  Filter by Task Name

### Returns

- `errors: array of unknown`

- `messages: array of string`

- `result: array of unknown`

- `success: boolean`

### Example

```http
curl https://api.cloudflare.com/client/v4/accounts/$ACCOUNT_ID/ai/models/search \
    -H "Authorization: Bearer $CLOUDFLARE_API_TOKEN"
```

#### Response

```json
{
  "errors": [
    {}
  ],
  "messages": [
    "string"
  ],
  "result": [
    {}
  ],
  "success": true
}
```

## Domain Types

### Model List Response

- `ModelListResponse = unknown`

# Schema

## Get Model Schema

**get** `/accounts/{account_id}/ai/models/schema`

Retrieves the input and output JSON schema definition for a Workers AI model.

### Path Parameters

- `account_id: string`

### Query Parameters

- `model: string`

  Model Name

### Returns

- `result: object { input, output }`

  - `input: object { additionalProperties, description, type }`

    - `additionalProperties: boolean`

    - `description: string`

    - `type: string`

  - `output: object { additionalProperties, description, type }`

    - `additionalProperties: boolean`

    - `description: string`

    - `type: string`

- `success: boolean`

### Example

```http
curl https://api.cloudflare.com/client/v4/accounts/$ACCOUNT_ID/ai/models/schema \
    -H "Authorization: Bearer $CLOUDFLARE_API_TOKEN"
```

#### Response

```json
{
  "result": {
    "input": {
      "additionalProperties": true,
      "description": "JSON Schema definition for the model's input parameters",
      "type": "object"
    },
    "output": {
      "additionalProperties": true,
      "description": "JSON Schema definition for the model's output format",
      "type": "object"
    }
  },
  "success": true
}
```

## Domain Types

### Schema Get Response

- `SchemaGetResponse = object { input, output }`

  - `input: object { additionalProperties, description, type }`

    - `additionalProperties: boolean`

    - `description: string`

    - `type: string`

  - `output: object { additionalProperties, description, type }`

    - `additionalProperties: boolean`

    - `description: string`

    - `type: string`

# To Markdown

## Convert Files into Markdown

**post** `/accounts/{account_id}/ai/tomarkdown`

Converts uploaded files into Markdown format using Workers AI.

### Path Parameters

- `account_id: string`

### Returns

- `result: array of object { data, format, mimeType, 2 more }`

  - `data: string`

  - `format: string`

  - `mimeType: string`

  - `name: string`

  - `tokens: string`

- `success: boolean`

### Example

```http
curl https://api.cloudflare.com/client/v4/accounts/$ACCOUNT_ID/ai/tomarkdown \
    -H 'Content-Type: multipart/form-data' \
    -H "Authorization: Bearer $CLOUDFLARE_API_TOKEN" \
    -F file='{"files":["Example data"]}'
```

#### Response

```json
{
  "result": [
    {
      "data": "data",
      "format": "format",
      "mimeType": "mimeType",
      "name": "name",
      "tokens": "tokens"
    }
  ],
  "success": true
}
```

## Get all converted formats supported

**get** `/accounts/{account_id}/ai/tomarkdown/supported`

Lists all file formats supported for conversion to Markdown.

### Path Parameters

- `account_id: string`

### Returns

- `result: array of object { extension, mimeType }`

  - `extension: string`

  - `mimeType: string`

- `success: boolean`

### Example

```http
curl https://api.cloudflare.com/client/v4/accounts/$ACCOUNT_ID/ai/tomarkdown/supported \
    -H "Authorization: Bearer $CLOUDFLARE_API_TOKEN"
```

#### Response

```json
{
  "result": [
    {
      "extension": "extension",
      "mimeType": "mimeType"
    }
  ],
  "success": true
}
```

## Domain Types

### To Markdown Transform Response

- `ToMarkdownTransformResponse = object { data, format, mimeType, 2 more }`

  - `data: string`

  - `format: string`

  - `mimeType: string`

  - `name: string`

  - `tokens: string`

### To Markdown Supported Response

- `ToMarkdownSupportedResponse = object { extension, mimeType }`

  - `extension: string`

  - `mimeType: string`