# CF AI Hello World Sample (Cloudflare Workers AI)
This sample demonstrates how to use the [Cloudflare Workers AI](https://developers.cloudflare.com/workers-ai/) plugin for Genkit.
## Prerequisites
1. A [Cloudflare account](https://dash.cloudflare.com/) with Workers AI enabled
2. Your Cloudflare Account ID
3. An API token with Workers AI permissions
## Setup
1. Export your credentials:
```bash
export CLOUDFLARE_ACCOUNT_ID=your_account_id
export CLOUDFLARE_API_TOKEN=your_api_token
```
2. Run the sample:
```bash
./run.sh
```
## Features Demonstrated
### Text Generation
The `say_hello` flow uses Llama 3.1 8B to generate a friendly greeting:
```python
response = await ai.generate(prompt=f'Say hello to {input.name}!')
```
### Streaming
The `streaming_demo` flow demonstrates token-by-token streaming:
```python
async for chunk in ai.generate_stream(prompt='Tell me a short joke.'):
print(chunk.text, end='')
```
### Tool Calling
The `tool_demo` flow shows how to use tools with Cloudflare models:
```python
@ai.tool()
async def get_weather(location: str) -> str:
return f"The weather in {location} is sunny, 72°F."
```
### Embeddings
The `embedding_demo` flow generates text embeddings:
```python
embeddings = await ai.embed(embedder=bge_base_en, documents=['Hello world'])
```
## Testing
After starting the sample, open the Genkit DevUI at http://localhost:4000 and:
1. **say_hello**: Enter a name to get a personalized greeting
2. **streaming_demo**: Watch tokens stream in real-time
3. **tool_demo**: See tool calling in action
4. **embedding_demo**: Generate and view embedding vectors
## Supported Models
### Text Generation
- `@cf/meta/llama-3.1-8b-instruct` - Default model
- `@cf/meta/llama-3.3-70b-instruct-fp8-fast` - Larger, more capable
- `@hf/mistral/mistral-7b-instruct-v0.2` - Mistral alternative (HuggingFace hosted)
### Embeddings
- `@cf/baai/bge-base-en-v1.5` - 768 dimensions
- `@cf/baai/bge-large-en-v1.5` - 1024 dimensions
- `@cf/baai/bge-small-en-v1.5` - 384 dimensions
