add readme and sample usage

crates/hermesllm/README.md

@@ -0,0 +1,63 @@
+# hermesllm
+
+A Rust library for translating LLM (Large Language Model) API requests and responses between Mistral, Groq, Gemini, Deepseek, OpenAI, and other provider-compliant formats.
+
+## Features
+
+- Unified types for chat completions and model metadata across multiple LLM providers
+- Builder-pattern API for constructing requests in an idiomatic Rust style
+- Easy conversion between provider formats
+- Streaming and non-streaming response support
+
+## Supported Providers
+
+- Mistral
+- Deepseek
+- Groq
+- Gemini
+- OpenAI
+- Claude
+- Github
+
+## Installation
+
+Add the following to your `Cargo.toml`:
+
+```toml
+[dependencies]
+hermesllm = { git = "https://github.com/katanemo/archgw", subdir = "crates/hermesllm" }
+```
+
+_Replace the path with the appropriate location if using as a workspace member or published crate._
+
+## Usage
+
+Construct a chat completion request using the builder pattern:
+
+```rust
+use hermesllm::Provider;
+use hermesllm::providers::openai::types::ChatCompletionsRequest;
+
+let request = ChatCompletionsRequest::builder("gpt-3.5-turbo", vec![Message::new("Hi".to_string())])
+    .build()
+    .expect("Failed to build OpenAIRequest");
+
+// Convert to bytes for a specific provider
+let bytes = request.to_bytes(Provider::OpenAI)?;
+```
+
+## API Overview
+
+- `Provider`: Enum listing all supported LLM providers.
+- `ChatCompletionsRequest`: Builder-pattern struct for creating chat completion requests.
+- `ChatCompletionsResponse`: Struct for parsing responses.
+- Streaming support via `SseChatCompletionIter`.
+- Error handling via `OpenAIError`.
+
+## Contributing
+
+Contributions are welcome! Please open issues or pull requests for bug fixes, new features, or provider integrations.
+
+## License
+
+This project is licensed under the terms of the [MIT License](../LICENSE).

crates/hermesllm/src/lib.rs

@@ -49,21 +49,22 @@ impl Display for Provider {
 
 #[cfg(test)]
 mod tests {
-    use crate::providers::openai::types::ChatCompletionsRequest;
+    use crate::providers::openai::types::{ChatCompletionsRequest, Message};
 
     #[test]
     fn openai_builder() {
-        let request = ChatCompletionsRequest::builder("gpt-3.5-turbo", vec![])
-            .temperature(0.7)
-            .top_p(0.9)
-            .n(1)
-            .max_tokens(100)
-            .stream(false)
-            .stop(vec!["\n".to_string()])
-            .presence_penalty(0.0)
-            .frequency_penalty(0.0)
-            .build()
-            .expect("Failed to build OpenAIRequest");
+        let request =
+            ChatCompletionsRequest::builder("gpt-3.5-turbo", vec![Message::new("Hi".to_string())])
+                .temperature(0.7)
+                .top_p(0.9)
+                .n(1)
+                .max_tokens(100)
+                .stream(false)
+                .stop(vec!["\n".to_string()])
+                .presence_penalty(0.0)
+                .frequency_penalty(0.0)
+                .build()
+                .expect("Failed to build OpenAIRequest");
 
         assert_eq!(request.model, "gpt-3.5-turbo");
         assert_eq!(request.temperature, Some(0.7));

crates/hermesllm/src/providers/openai/types.rs

@@ -80,6 +80,15 @@ pub struct Message {
     pub content: Option<ContentType>,
 }
 
+impl Message {
+    pub fn new(content: String) -> Self {
+        Self {
+            role: "user".to_string(),
+            content: Some(ContentType::Text(content)),
+        }
+    }
+}
+
 #[derive(Debug, Clone, Serialize, Deserialize)]
 pub struct StreamOptions {
     pub include_usage: bool,