Together.ai AI SDK Integration

Integrate Together.ai's language models into your AI SDK applications. This skill provides access to models hosted on the Together.ai platform, including Meta-Llama and other open-source models.

What This Skill Does

This skill helps you set up and use the Together.ai provider for the AI SDK, enabling you to:

Instructions

When a user requests Together.ai integration with AI SDK, follow these steps:

1. Install Dependencies

Install the Together.ai provider package:

```bash

npm install @ai-sdk/togetherai

```

If the user doesn't have the base AI SDK installed:

```bash

npm install ai

```

2. Set Up Environment Variables

Ensure the user has a Together.ai API key configured. Add to their `.env` file if not present:

```

TOGETHER_API_KEY=your_together_ai_api_key_here

```

Remind the user to:

3. Import and Initialize Provider

Create or update the relevant file with the Together.ai provider:

```typescript

import { togetherai } from '@ai-sdk/togetherai';

```

4. Generate Text Example

Provide a working example based on the user's use case. Default example:

```typescript

import { togetherai } from '@ai-sdk/togetherai';

import { generateText } from 'ai';

const { text } = await generateText({

model: togetherai('meta-llama/Meta-Llama-3.1-8B-Instruct-Turbo'),

prompt: 'Write a Python function that sorts a list:',

});

console.log(text);

```

5. Model Selection

Common Together.ai models to suggest:

Adjust model selection based on user requirements (speed vs capability, cost considerations).

6. Additional AI SDK Features

If the user needs more than basic text generation, mention these AI SDK capabilities:

Provide code examples only if the user specifically requests these features.

Example Usage Scenarios

**Scenario 1: Simple text generation**

```typescript

import { togetherai } from '@ai-sdk/togetherai';

import { generateText } from 'ai';

const result = await generateText({

model: togetherai('meta-llama/Meta-Llama-3.1-8B-Instruct-Turbo'),

prompt: 'Explain quantum computing in simple terms',

});

```

**Scenario 2: Streaming responses**

```typescript

import { togetherai } from '@ai-sdk/togetherai';

import { streamText } from 'ai';

const { textStream } = await streamText({

model: togetherai('meta-llama/Meta-Llama-3.1-70B-Instruct-Turbo'),

prompt: 'Write a story about AI',

});

for await (const chunk of textStream) {

process.stdout.write(chunk);

}

```

**Scenario 3: Structured output**

```typescript

import { togetherai } from '@ai-sdk/togetherai';

import { generateObject } from 'ai';

import { z } from 'zod';

const result = await generateObject({

model: togetherai('meta-llama/Meta-Llama-3.1-8B-Instruct-Turbo'),

schema: z.object({

recipe: z.object({

name: z.string(),

ingredients: z.array(z.string()),

steps: z.array(z.string()),

}),

}),

prompt: 'Generate a recipe for chocolate chip cookies',

});

```

Important Notes

Constraints