autoagents-llm 0.4.0

//! Groq API client implementation for chat functionality.
//!
//! This module provides integration with Groq's LLM models through their API.

use crate::builder::LLMBuilder;
use crate::http::ensure_success;
use crate::{
    LLMProvider,
    builder::LLMBackend,
    chat::{StructuredOutputFormat, ToolChoice},
    completion::{CompletionProvider, CompletionRequest, CompletionResponse},
    embedding::EmbeddingProvider,
    error::LLMError,
    models::{ModelListRequest, ModelListResponse, ModelsProvider, StandardModelListResponse},
    providers::openai_compatible::{OpenAICompatibleProvider, OpenAIProviderConfig},
};
use async_trait::async_trait;
use std::sync::Arc;

/// Groq configuration for the generic provider
pub struct GroqConfig;

impl OpenAIProviderConfig for GroqConfig {
    const PROVIDER_NAME: &'static str = "Groq";
    const DEFAULT_BASE_URL: &'static str = "https://api.groq.com/openai/v1/";
    const DEFAULT_MODEL: &'static str = "llama3-8b-8192";
    const SUPPORTS_REASONING_EFFORT: bool = false;
    const SUPPORTS_STRUCTURED_OUTPUT: bool = true;
    const SUPPORTS_PARALLEL_TOOL_CALLS: bool = false;
}

pub type Groq = OpenAICompatibleProvider<GroqConfig>;

// TODO: for groq usage goes in .x_groq.usage...
// /// Streaming response structures
// #[derive(Deserialize, Debug)]
// pub struct GroqChatStreamChunk {
//     pub choices: Vec<ChatStreamChoice>,
//     pub x_groq: Option<GroqMetadata>,
// }

// /// Streaming response structures
// #[derive(Deserialize, Debug)]
// pub struct GroqMetadata {
//     pub usage: Option<Usage>,
// }

impl Groq {
    /// Creates a new Groq client with the specified configuration.
    #[allow(clippy::too_many_arguments)]
    pub fn with_config(
        api_key: impl Into<String>,
        base_url: Option<String>,
        model: Option<String>,
        max_tokens: Option<u32>,
        temperature: Option<f32>,
        timeout_seconds: Option<u64>,
        top_p: Option<f32>,
        top_k: Option<u32>,
        tool_choice: Option<ToolChoice>,
        _embedding_encoding_format: Option<String>,
        _embedding_dimensions: Option<u32>,
        reasoning_effort: Option<String>,
        parallel_tool_calls: Option<bool>,
        normalize_response: Option<bool>,
        extra_body: Option<serde_json::Value>,
    ) -> Self {
        OpenAICompatibleProvider::<GroqConfig>::new(
            api_key,
            base_url,
            model,
            max_tokens,
            temperature,
            timeout_seconds,
            top_p,
            top_k,
            tool_choice,
            reasoning_effort,
            None, // voice - not supported by Groq,
            extra_body,
            parallel_tool_calls,
            normalize_response,
            None, // embedding_encoding_format - not supported by Groq
            None, // embedding_dimensions - not supported by Groq
        )
    }
}

impl LLMProvider for Groq {}

impl crate::HasConfig for Groq {
    type Config = crate::NoConfig;
}

#[async_trait]
impl CompletionProvider for Groq {
    async fn complete(
        &self,
        _req: &CompletionRequest,
        _json_schema: Option<StructuredOutputFormat>,
    ) -> Result<CompletionResponse, LLMError> {
        Ok(CompletionResponse {
            text: "Groq completion not implemented.".into(),
        })
    }
}

#[async_trait]
impl EmbeddingProvider for Groq {
    async fn embed(&self, _text: Vec<String>) -> Result<Vec<Vec<f32>>, LLMError> {
        Err(LLMError::ProviderError(
            "Embedding not supported".to_string(),
        ))
    }
}

#[async_trait]
impl ModelsProvider for Groq {
    async fn list_models(
        &self,
        _request: Option<&ModelListRequest>,
    ) -> Result<Box<dyn ModelListResponse>, LLMError> {
        if self.api_key.is_empty() {
            return Err(LLMError::missing_api_key(
                "Missing Groq API key".to_string(),
            ));
        }

        let url = format!("{}/models", GroqConfig::DEFAULT_BASE_URL);

        let resp = self
            .client
            .get(&url)
            .bearer_auth(&self.api_key)
            .send()
            .await?;
        let resp = ensure_success(resp, "Groq").await?;

        let result = StandardModelListResponse {
            inner: resp.json().await?,
            backend: LLMBackend::Groq,
        };
        Ok(Box::new(result))
    }
}

impl LLMBuilder<Groq> {
    pub fn build(self) -> Result<Arc<Groq>, LLMError> {
        let api_key = self
            .api_key
            .ok_or_else(|| LLMError::invalid_request("No API key provided for Groq".to_string()))?;

        let groq = Groq::with_config(
            api_key,
            self.base_url,
            self.model,
            self.max_tokens,
            self.temperature,
            self.timeout_seconds,
            self.top_p,
            self.top_k,
            self.tool_choice,
            None, // embedding_encoding_format
            None, // embedding_dimensions
            self.reasoning_effort,
            self.enable_parallel_tool_use,
            self.normalize_response,
            self.extra_body,
        );

        Ok(Arc::new(groq))
    }
}

#[cfg(test)]
mod tests {
    use super::*;
    use crate::builder::LLMBuilder;
    use crate::completion::CompletionRequest;

    #[test]
    fn test_with_config_defaults() {
        let provider = Groq::with_config(
            "key",
            None,
            None,
            Some(200),
            Some(0.5),
            Some(12),
            None,
            None,
            None,
            None,
            None,
            None,
            None,
            None,
            None,
        );

        assert_eq!(provider.api_key, "key");
        assert_eq!(provider.model, GroqConfig::DEFAULT_MODEL);
        assert_eq!(provider.max_tokens, Some(200));
        assert_eq!(provider.temperature, Some(0.5));
        assert_eq!(provider.timeout_seconds, 12);
    }

    #[tokio::test]
    async fn test_list_models_missing_key() {
        let provider = Groq::with_config(
            "", None, None, None, None, None, None, None, None, None, None, None, None, None, None,
        );
        let err = provider.list_models(None).await.unwrap_err();
        assert!(err.to_string().contains("Missing Groq API key"));
    }

    #[tokio::test]
    async fn test_complete_returns_placeholder() {
        let provider = Groq::with_config(
            "key", None, None, None, None, None, None, None, None, None, None, None, None, None,
            None,
        );
        let response = provider
            .complete(
                &CompletionRequest {
                    prompt: "hi".to_string(),
                    max_tokens: None,
                    temperature: None,
                },
                None,
            )
            .await
            .unwrap();
        assert!(response.text.contains("Groq completion not implemented"));
    }

    #[tokio::test]
    async fn test_embed_not_supported() {
        let provider = Groq::with_config(
            "key", None, None, None, None, None, None, None, None, None, None, None, None, None,
            None,
        );
        let err = provider.embed(vec!["hello".to_string()]).await.unwrap_err();
        assert!(err.to_string().contains("Embedding not supported"));
    }

    #[test]
    fn test_builder_requires_api_key() {
        let result = LLMBuilder::<Groq>::new().build();
        assert!(result.is_err());
        let err = result.err().unwrap();
        assert!(err.to_string().contains("No API key provided for Groq"));
    }
}