evolve-llm 0.2.0

Minimal LLM client (Anthropic Haiku + Ollama) for occasional challenger generation
Documentation 
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
42
43
44

//! The `LlmClient` trait and its shared types.

use crate::error::LlmError;
use async_trait::async_trait;

/// Token usage reported by the provider. Both fields are 0 if unknown.
#[derive(Debug, Clone, Copy, Default, PartialEq, Eq)]
pub struct TokenUsage {
    /// Input tokens billed.
    pub input: u32,
    /// Output tokens billed.
    pub output: u32,
}

/// One completion response.
#[derive(Debug, Clone, PartialEq, Eq)]
pub struct CompletionResult {
    /// Assistant text.
    pub text: String,
    /// Token usage.
    pub usage: TokenUsage,
}

/// Shared interface for LLM clients.
#[async_trait]
pub trait LlmClient: Send + Sync + std::fmt::Debug {
    /// Run a single non-streaming completion.
    async fn complete(&self, prompt: &str, max_tokens: u32) -> Result<CompletionResult, LlmError>;

    /// Stable identifier used by the cost tracker price table.
    fn model_id(&self) -> &str;
}

#[cfg(test)]
mod tests {
    use super::*;

    #[test]
    fn token_usage_default_is_zero() {
        let u = TokenUsage::default();
        assert_eq!(u.input, 0);
        assert_eq!(u.output, 0);
    }
}