> ## Documentation Index
> Fetch the complete documentation index at: https://docs.corsair.dev/llms.txt
> Use this file to discover all available pages before exploring further.

# API

> API reference for Ollama: every `ollama.api.*` operation with input and output types.

Every `ollama.api.*` operation is listed below with parameter shapes and return types from the plugin Zod schemas.

<Info>
  **New to Corsair?** See [API access](/concepts/api), [authentication](/concepts/auth), and [error handling](/concepts/error-handling).
</Info>

## Chat

### chat

`chat.chat`

Tool to send a chat message with conversation history to Ollama. Use when you need to have a multi-turn conversation with an LLM model.

**Risk:** `write`

```ts theme={null}
await corsair.ollama.api.chat.chat({});
```

**Input**

| Name         | Type               | Required | Description                                             |
| ------------ | ------------------ | -------- | ------------------------------------------------------- |
| `model`      | `string`           | Yes      | Name of the model to use for chat                       |
| `messages`   | `object[]`         | Yes      | Conversation history                                    |
| `tools`      | `any[]`            | No       | List of tools/functions available to the model          |
| `format`     | `object`           | No       | Output format, e.g. "json" or a JSON schema             |
| `options`    | `object`           | No       | Model configuration options (temperature, top\_p, etc.) |
| `stream`     | `boolean`          | No       | Whether to stream responses (default false)             |
| `keep_alive` | `string \| number` | No       | Duration to keep the model loaded in memory             |

<AccordionGroup>
  <Accordion title="messages full type">
    ```ts theme={null}
    {
      role: string,
      content: string,
      images?: string[],
      tool_calls?: any[]
    }[]
    ```
  </Accordion>

  <Accordion title="format full type">
    ```ts theme={null}
    string | {
    }
    ```
  </Accordion>

  <Accordion title="options full type">
    ```ts theme={null}
    {
    }
    ```
  </Accordion>
</AccordionGroup>

**Output**

| Name                   | Type      | Required | Description |
| ---------------------- | --------- | -------- | ----------- |
| `model`                | `string`  | Yes      | —           |
| `created_at`           | `string`  | Yes      | —           |
| `message`              | `object`  | Yes      | —           |
| `done`                 | `boolean` | Yes      | —           |
| `total_duration`       | `number`  | No       | —           |
| `load_duration`        | `number`  | No       | —           |
| `prompt_eval_count`    | `number`  | No       | —           |
| `prompt_eval_duration` | `number`  | No       | —           |
| `eval_count`           | `number`  | No       | —           |
| `eval_duration`        | `number`  | No       | —           |

<AccordionGroup>
  <Accordion title="message full type">
    ```ts theme={null}
    {
      role: string,
      content: string,
      images?: string[],
      tool_calls?: any[]
    }
    ```
  </Accordion>
</AccordionGroup>

***

### generate

`chat.generate`

Tool to generate text responses from Ollama models with optional raw mode. Use raw=true to bypass prompt templating when you need full control over the prompt for debugging or custom processing. Note that raw mode will not return a context.

**Risk:** `write`

```ts theme={null}
await corsair.ollama.api.chat.generate({});
```

**Input**

| Name         | Type               | Required | Description                                 |
| ------------ | ------------------ | -------- | ------------------------------------------- |
| `model`      | `string`           | Yes      | Name of the model to generate text with     |
| `prompt`     | `string`           | No       | The prompt to generate a response for       |
| `suffix`     | `string`           | No       | Text after the insertion point              |
| `images`     | `string[]`         | No       | Base64-encoded images for multimodal models |
| `format`     | `object`           | No       | Format of the response                      |
| `options`    | `object`           | No       | Model configuration options                 |
| `system`     | `string`           | No       | System message to override model default    |
| `template`   | `string`           | No       | Prompt template to override model default   |
| `stream`     | `boolean`          | No       | Whether to stream response                  |
| `raw`        | `boolean`          | No       | Bypass prompt template when true            |
| `keep_alive` | `string \| number` | No       | Duration to keep the model loaded           |

<AccordionGroup>
  <Accordion title="format full type">
    ```ts theme={null}
    string | {
    }
    ```
  </Accordion>

  <Accordion title="options full type">
    ```ts theme={null}
    {
    }
    ```
  </Accordion>
</AccordionGroup>

**Output**

| Name                   | Type       | Required | Description |
| ---------------------- | ---------- | -------- | ----------- |
| `model`                | `string`   | Yes      | —           |
| `created_at`           | `string`   | Yes      | —           |
| `response`             | `string`   | Yes      | —           |
| `done`                 | `boolean`  | Yes      | —           |
| `context`              | `number[]` | No       | —           |
| `total_duration`       | `number`   | No       | —           |
| `load_duration`        | `number`   | No       | —           |
| `prompt_eval_count`    | `number`   | No       | —           |
| `prompt_eval_duration` | `number`   | No       | —           |
| `eval_count`           | `number`   | No       | —           |
| `eval_duration`        | `number`   | No       | —           |

***

## Models

### listModels

`models.listModels`

Tool to list all available Ollama models and their details. Use when you need to fetch installed models with metadata including name, size, last modified timestamp, digest, and format information.

**Risk:** `read`

```ts theme={null}
await corsair.ollama.api.models.listModels({});
```

**Input:** *empty object*

**Output**

| Name     | Type       | Required | Description |
| -------- | ---------- | -------- | ----------- |
| `models` | `object[]` | Yes      | —           |

<AccordionGroup>
  <Accordion title="models full type">
    ```ts theme={null}
    {
      name: string,
      model?: string,
      modified_at?: string,
      size?: number,
      digest?: string,
      details?: {
        parent_model?: string,
        format?: string,
        family?: string,
        families?: string[],
        parameter_size?: string,
        quantization_level?: string
      }
    }[]
    ```
  </Accordion>
</AccordionGroup>

***

### showModel

`models.showModel`

Tool to show comprehensive information about an Ollama model. Use when you need to retrieve model details, parameters, template, license, or system prompt.

**Risk:** `read`

```ts theme={null}
await corsair.ollama.api.models.showModel({});
```

**Input**

| Name       | Type     | Required | Description                               |
| ---------- | -------- | -------- | ----------------------------------------- |
| `model`    | `string` | Yes      | Name of the model to show information for |
| `system`   | `string` | No       | System prompt override                    |
| `template` | `string` | No       | Template override                         |
| `options`  | `object` | No       | Options override                          |

<AccordionGroup>
  <Accordion title="options full type">
    ```ts theme={null}
    {
    }
    ```
  </Accordion>
</AccordionGroup>

**Output**

| Name          | Type     | Required | Description |
| ------------- | -------- | -------- | ----------- |
| `modelfile`   | `string` | No       | —           |
| `parameters`  | `string` | No       | —           |
| `template`    | `string` | No       | —           |
| `details`     | `object` | No       | —           |
| `model_info`  | `object` | No       | —           |
| `modified_at` | `string` | No       | —           |
| `license`     | `string` | No       | —           |

<AccordionGroup>
  <Accordion title="details full type">
    ```ts theme={null}
    {
      parent_model?: string,
      format?: string,
      family?: string,
      families?: string[],
      parameter_size?: string,
      quantization_level?: string
    }
    ```
  </Accordion>

  <Accordion title="model_info full type">
    ```ts theme={null}
    {
    }
    ```
  </Accordion>
</AccordionGroup>

***

### version

`models.version`

Tool to get the version of Ollama running locally. Use to check which version of Ollama is currently installed.

**Risk:** `read`

```ts theme={null}
await corsair.ollama.api.models.version({});
```

**Input:** *empty object*

**Output**

| Name      | Type     | Required | Description |
| --------- | -------- | -------- | ----------- |
| `version` | `string` | Yes      | —           |

***

## Openai

### createOpenAiChatCompletion

`openai.createOpenAiChatCompletion`

Tool to create OpenAI-compatible chat completions using Ollama models. Use when you need conversational AI responses with OpenAI API format compatibility.

**Risk:** `write`

```ts theme={null}
await corsair.ollama.api.openai.createOpenAiChatCompletion({});
```

**Input**

| Name                | Type                 | Required | Description                             |
| ------------------- | -------------------- | -------- | --------------------------------------- |
| `model`             | `string`             | Yes      | ID of the model to use                  |
| `messages`          | `object[]`           | Yes      | Messages in the conversation            |
| `temperature`       | `number`             | No       | Sampling temperature                    |
| `top_p`             | `number`             | No       | Nucleus sampling probability            |
| `n`                 | `number`             | No       | Number of completions to generate       |
| `stream`            | `boolean`            | No       | Whether to stream responses             |
| `stop`              | `string \| string[]` | No       | Stop sequences                          |
| `max_tokens`        | `number`             | No       | Maximum tokens to generate              |
| `presence_penalty`  | `number`             | No       | Presence penalty                        |
| `frequency_penalty` | `number`             | No       | Frequency penalty                       |
| `user`              | `string`             | No       | Unique identifier representing end-user |

<AccordionGroup>
  <Accordion title="messages full type">
    ```ts theme={null}
    {
      role: string,
      content: string
    }[]
    ```
  </Accordion>
</AccordionGroup>

**Output**

| Name      | Type       | Required | Description |
| --------- | ---------- | -------- | ----------- |
| `id`      | `string`   | Yes      | —           |
| `object`  | `string`   | Yes      | —           |
| `created` | `number`   | Yes      | —           |
| `model`   | `string`   | Yes      | —           |
| `choices` | `object[]` | Yes      | —           |
| `usage`   | `object`   | No       | —           |

<AccordionGroup>
  <Accordion title="choices full type">
    ```ts theme={null}
    {
      index: number,
      message: {
        role: string,
        content: string
      },
      finish_reason?: string | null
    }[]
    ```
  </Accordion>

  <Accordion title="usage full type">
    ```ts theme={null}
    {
      prompt_tokens: number,
      completion_tokens: number,
      total_tokens: number
    }
    ```
  </Accordion>
</AccordionGroup>

***

### createOpenAiCompletion

`openai.createOpenAiCompletion`

Tool to create OpenAI-compatible text completions using Ollama models. Use when you need text generation with OpenAI API format compatibility beyond chat-based interactions.

**Risk:** `write`

```ts theme={null}
await corsair.ollama.api.openai.createOpenAiCompletion({});
```

**Input**

| Name                | Type                 | Required | Description                                |
| ------------------- | -------------------- | -------- | ------------------------------------------ |
| `model`             | `string`             | Yes      | ID of the model to use                     |
| `prompt`            | `string \| string[]` | Yes      | Prompt text to complete                    |
| `suffix`            | `string`             | No       | Suffix to insert                           |
| `max_tokens`        | `number`             | No       | Maximum tokens to generate                 |
| `temperature`       | `number`             | No       | Sampling temperature                       |
| `top_p`             | `number`             | No       | Nucleus sampling probability               |
| `n`                 | `number`             | No       | Number of completions to generate          |
| `stream`            | `boolean`            | No       | Whether to stream responses                |
| `logprobs`          | `number`             | No       | Include log probabilities                  |
| `echo`              | `boolean`            | No       | Echo prompt in completion                  |
| `stop`              | `string \| string[]` | No       | Stop sequences                             |
| `presence_penalty`  | `number`             | No       | Presence penalty                           |
| `frequency_penalty` | `number`             | No       | Frequency penalty                          |
| `best_of`           | `number`             | No       | Generates best\_of completions server-side |
| `user`              | `string`             | No       | Unique identifier representing end-user    |

**Output**

| Name      | Type       | Required | Description |
| --------- | ---------- | -------- | ----------- |
| `id`      | `string`   | Yes      | —           |
| `object`  | `string`   | Yes      | —           |
| `created` | `number`   | Yes      | —           |
| `model`   | `string`   | Yes      | —           |
| `choices` | `object[]` | Yes      | —           |
| `usage`   | `object`   | No       | —           |

<AccordionGroup>
  <Accordion title="choices full type">
    ```ts theme={null}
    {
      index: number,
      text: string,
      logprobs?: any | null,
      finish_reason?: string | null
    }[]
    ```
  </Accordion>

  <Accordion title="usage full type">
    ```ts theme={null}
    {
      prompt_tokens: number,
      completion_tokens: number,
      total_tokens: number
    }
    ```
  </Accordion>
</AccordionGroup>

***

### listOpenAiModels

`openai.listOpenAiModels`

Tool to list available models using OpenAI-compatible API format. Use when you need to retrieve locally available Ollama models with metadata following OpenAI's model list format.

**Risk:** `read`

```ts theme={null}
await corsair.ollama.api.openai.listOpenAiModels({});
```

**Input:** *empty object*

**Output**

| Name     | Type       | Required | Description |
| -------- | ---------- | -------- | ----------- |
| `object` | `string`   | Yes      | —           |
| `data`   | `object[]` | Yes      | —           |

<AccordionGroup>
  <Accordion title="data full type">
    ```ts theme={null}
    {
      id: string,
      object: string,
      created?: number,
      owned_by?: string
    }[]
    ```
  </Accordion>
</AccordionGroup>

***