# Ollama

> OLLamaLLMService streams chat completions from Ollama through its OpenAI-compatible API.

Source: https://polvallverdu.github.io/pipedog/docs/services/llm/ollama/



{/* Generated by scripts/generate-providers.ts from data/providers.ts. */}

## Overview [#overview]

Ollama runs open models on your own machine, behind an OpenAI-compatible API.

* **`OLLamaLLMService`**: an `OpenAILLMService` subclass with Ollama's base URL and default
  model.

<Cards>
  <Card title="Upstream page" href="https://docs.pipecat.ai/api-reference/server/services/llm/ollama">
    Pipecat's documentation for the same service.
  </Card>

  <Card title="Source" href="https://github.com/polvallverdu/pipedog/tree/main/packages/ollama/src">
    `packages/ollama/src/llm.ts`.
  </Card>

  <Card title="Ollama library" href="https://ollama.com/library">
    The models you can pull.
  </Card>
</Cards>

## Installation [#installation]

```bash
bun add @pipedog/ollama
```

It depends on `@pipedog/openai`, which is installed with it.

```ts
import { OLLamaLLMService } from "@pipedog/ollama/llm";
```

## Prerequisites [#prerequisites]

* [Ollama](https://ollama.com/download) running locally, with a model pulled (`ollama pull llama2`). No API key is needed; `apiKey` defaults to `"ollama"`, which a local Ollama ignores.

## OLLamaLLMService [#ollamallmservice]

### Configuration [#configuration]

| Option | Type | Required | Description |
|---|---|---|---|
| `name` | `string \| null \| undefined` | no | Custom name; defaults to `ClassName#count`. |
| `taskManager` | `BaseTaskManager \| null \| undefined` | no | Task manager for handling tasks. |
| `baseUrl` | `string \| null \| undefined` | no | The base URL for the OLLama API endpoint. Defaults to `"http://localhost:11434/v1"`. |
| `apiKey` | `string \| null \| undefined` | no | API key for an Ollama server that requires one, such as one behind an authenticating proxy. A local Ollama ignores it. Defaults to `"ollama"`. |
| `settings` | `OllamaLLMSettings \| null \| undefined` | no | Runtime-updatable settings; values override the built-in defaults. |
| `enableDirectMode` | `boolean \| undefined` | no | Process frames immediately instead of using internal queues. |
| `metrics` | `FrameProcessorMetrics \| null \| undefined` | no | Optional metrics collector for this processor. |
| `runInParallel` | `boolean \| undefined` | no | Whether to run function calls in parallel or sequentially. Defaults to true. |
| `groupParallelTools` | `boolean \| undefined` | no | Whether to group parallel function calls so the LLM is triggered exactly once after all calls in the batch complete. When false, each function call result triggers the LLM independently as it arrives. Defaults to true. |
| `functionCallTimeoutSecs` | `number \| null \| undefined` | no | Optional timeout in seconds for function calls. A call that runs past it is cancelled: its handler is cancelled (it gets a `CancelledError`), the call is settled as cancelled, and inference runs so the LLM can report that it didn't complete. |
| `fileResolver` | `FileResolver \| null \| undefined` | no | Resolves file URLs in the context that this provider can't fetch itself (see `FileResolver`). Pass one configured with a `fileStorage` to resolve uploaded-file URLs (`new FileResolver({ fileStorage: runnerArgs.fileStorage })` when a runner hosts the bot), and share one instance across services that may consume the same files (e.g. services switched between mid-session) so downloads are reused. Defaults to a resolver that handles `data:` and `http(s)` URLs only. |
| `organization` | `string \| null \| undefined` | no | OpenAI organization ID. |
| `project` | `string \| null \| undefined` | no | OpenAI project ID. |
| `defaultHeaders` | `Record<string, string> \| null \| undefined` | no | Additional HTTP headers. |
| `serviceTier` | `string \| null \| undefined` | no | Service tier to use (e.g., "auto", "flex", "priority"). |
| `retryTimeoutSecs` | `number \| null \| undefined` | no | How long an inference may go without producing output before it is abandoned and re-issued, when `retryOnTimeout` is set. Defaults to 5.0 seconds. |
| `retryOnTimeout` | `boolean \| null \| undefined` | no | Whether to re-issue the request once if the first attempt produces no output within `retryTimeoutSecs`. The retry is unbounded. |

The other options (`organization`, `project`, `defaultHeaders`, `serviceTier`,
`retryOnTimeout`, `retryTimeoutSecs`, function calling) are `OpenAILLMService`'s. See
[OpenAI](https://polvallverdu.github.io/pipedog/docs/services/llm/openai/#configuration).

|               |                                                               |
| ------------- | ------------------------------------------------------------- |
| Base URL      | `http://localhost:11434/v1`                                   |
| Default model | `llama2`                                                      |
| Models to try | `llama2`, `mistral`. See the provider's catalog for the rest. |

### Settings [#settings]

Runtime-updatable settings, passed as `settings: new OLLamaLLMService.Settings({...})`
(`OllamaLLMSettings`). They are [OpenAI's settings](https://polvallverdu.github.io/pipedog/docs/services/llm/openai/#settings), with these
defaults:

| Setting | Type     | Default    | Description |
| ------- | -------- | ---------- | ----------- |
| `model` | `string` | `"llama2"` | Model name. |

Every other setting keeps OpenAI's default. Unset sampling settings are left out of the request, so
the provider's own defaults apply. `extra` carries provider-specific request fields.

### Usage [#usage]

```ts title="llm.ts"
import { LLMUpdateSettingsFrame } from "@pipedog/core/frames/frames";
import { OLLamaLLMService } from "@pipedog/ollama/llm";

const llm = new OLLamaLLMService({
  settings: new OLLamaLLMService.Settings({
    model: "llama2",
    systemInstruction: "You are a helpful assistant in a voice conversation.",
    temperature: 0.7,
  }),
});
```

An Ollama server on another host, behind a proxy that checks a key:

```ts
const remote = new OLLamaLLMService({
  baseUrl: "https://ollama.example.com/v1",
  apiKey: process.env.OLLAMA_PROXY_KEY ?? "",
  settings: new OLLamaLLMService.Settings({ model: "mistral" }),
});
```

Change settings mid-call with an `LLMUpdateSettingsFrame`. The next completion uses them:

```ts
// Queue it on the worker, or push it from a processor upstream of the service.
const update = new LLMUpdateSettingsFrame({
  delta: new OLLamaLLMService.Settings({ temperature: 0.2, maxTokens: 300 }),
});
```

### Notes [#notes]

* Runs a completion for each `LLMContextFrame`, between an `LLMFullResponseStartFrame` and an
  `LLMFullResponseEndFrame`, and streams the reply as `LLMTextFrame`s, as
  [`OpenAILLMService`](https://polvallverdu.github.io/pipedog/docs/services/llm/openai.md) does. Function calls, token usage metrics and
  `runInference` work the same way.
* "developer" messages are sent as "user" messages, since the API doesn't support that role (`supportsDeveloperRole` is `false`).
* `OLLamaLLMService` doesn't require an API key, unlike the other OpenAI-compatible services, and never falls back to `OPENAI_API_KEY`.
* Ollama seems to ignore "developer" messages, so they are sent as "user" messages.

## Runtime notes [#runtime-notes]

| Runtime            | Works | What it needs |
| ------------------ | ----- | ------------- |
| Bun                | ✅ yes | Nothing.      |
| Node.js            | ✅ yes | Nothing.      |
| Deno               | ✅ yes | Nothing.      |
| Cloudflare Workers | ✅ yes | Nothing.      |

Generated from `pipecat.runtimes` and `pipecat.compat` in `packages/ollama/package.json`: ✅ means the compat smoke test passes on that runtime.

* The client is the official `openai` JS SDK (through `@pipedog/openai`), which talks through
  `fetch`. It works on Bun, Node, Deno and Cloudflare Workers.
* The deprecated `model` constructor option is not ported. Use `settings`.
* Tested against a local OpenAI-compatible mock server, not against Ollama itself.