1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
// Import required modules from the LLM library
use llm::{
builder::{LLMBackend, LLMBuilder}, // Builder components for LLM configuration
chat::ChatMessage, // Chat-related structures
};
#[tokio::main]
async fn main() -> Result<(), Box<dyn std::error::Error>> {
// Retrieve Anthropic API key from environment variable or use fallback
let api_key = std::env::var("ANTHROPIC_API_KEY").unwrap_or("anthro-key".into());
// Initialize and configure the LLM client with validation
let llm = LLMBuilder::new()
.backend(LLMBackend::Anthropic) // Use Anthropic's Claude model
.model("claude-3-5-sonnet-20240620") // Specify model version
.api_key(api_key) // Set API credentials
.max_tokens(512) // Limit response length
.temperature(0.7) // Control response randomness
.validator(|resp| {
// Add JSON validation
serde_json::from_str::<serde_json::Value>(resp)
.map(|_| ())
.map_err(|e| e.to_string())
})
.validator_attempts(3) // Allow up to 3 retries on validation failure
.build()
.expect("Failed to build LLM (Phind)");
// Prepare the chat message requesting JSON output
let messages = vec![
ChatMessage::user().content("Please give me a valid JSON describing a cat named Garfield, color 'orange'. with format {name: string, color: string}. Return only the JSON, no other text").build(),
];
// Send chat request and handle the response
match llm.chat(&messages).await {
Ok(text) => println!("{text}"),
Err(e) => eprintln!("Chat error: {e}"),
}
Ok(())
}