refactor: move libs to lib dir

Signed-off-by: Neelay Shah <neelays@nvidia.com>
Co-authored-by: Ryan McCormick <rmccormick@nvidia.com>

llm/rust/triton-llm/src/model_card/model.rs → lib/llm/src/model_card/model.rs

@@ -37,7 +37,7 @@ use std::time::Duration;
 
 use derive_builder::Builder;
 
-use triton_distributed::slug::Slug;
+use triton_distributed_runtime::slug::Slug;
 
 pub const BUCKET_NAME: &str = "mdc";
 

llm/rust/triton-llm/src/preprocessor.rs → lib/llm/src/preprocessor.rs

@@ -35,11 +35,11 @@ use tracing;
 use crate::model_card::model::{ModelDeploymentCard, ModelInfo, TokenizerKind};
 use crate::preprocessor::prompt::OAIChatLikeRequest;
 
-use triton_distributed::engine::{AsyncEngine, AsyncEngineContextProvider, ResponseStream};
-use triton_distributed::pipeline::{
+use triton_distributed_runtime::engine::{AsyncEngine, AsyncEngineContextProvider, ResponseStream};
+use triton_distributed_runtime::pipeline::{
     async_trait, AsyncEngineContext, Error, ManyOut, Operator, SingleIn,
 };
-use triton_distributed::protocols::annotated::{Annotated, AnnotationsProvider};
+use triton_distributed_runtime::protocols::annotated::{Annotated, AnnotationsProvider};
 
 use crate::protocols::{
     common::{SamplingOptionsProvider, StopConditionsProvider},

llm/rust/triton-llm/src/protocols.rs → lib/llm/src/protocols.rs

@@ -36,7 +36,7 @@ pub type DataStream<T> = Pin<Box<dyn Stream<Item = T> + Send + Sync>>;
 // has become the common response envelope for triton-distributed.
 // We may want to move the original Annotated back here and has a Infallible conversion to the the
 // ResponseEnvelop in triton-distributed.
-pub use triton_distributed::protocols::annotated::Annotated;
+pub use triton_distributed_runtime::protocols::annotated::Annotated;
 
 /// The LLM responses have multiple different fields and nests of objects to get to the actual
 /// text completion returned. This trait can be applied to the `choice` level objects to extract

llm/rust/triton-llm/src/protocols/openai.rs → lib/llm/src/protocols/openai.rs

@@ -13,7 +13,7 @@
 // See the License for the specific language governing permissions and
 // limitations under the License.
 
-/// Forward openai_api_rs::v1 to triton_llm::protocols::openai::v1
+/// Forward openai_api_rs::v1 to triton_distributed_llm::protocols::openai::v1
 pub mod chat_completions;
 pub mod completions;
 pub mod models;

llm/rust/triton-llm/src/protocols/openai/chat_completions.rs → lib/llm/src/protocols/openai/chat_completions.rs

@@ -40,7 +40,8 @@ use super::{
     validate_logit_bias, ContentProvider, OpenAISamplingOptionsProvider,
     OpenAIStopConditionsProvider,
 };
-use triton_distributed::protocols::annotated::AnnotationsProvider;
+
+use triton_distributed_runtime::protocols::annotated::AnnotationsProvider;
 
 /// Request object which is used to generate chat completions.
 #[derive(Serialize, Deserialize, Builder, Validate, Debug, Clone)]
@@ -187,7 +188,7 @@ pub struct ChatCompletionRequest {
     /// in the vocabulary of the model.
     ///
     /// ```
-    /// use triton_llm::protocols::openai::completions::CompletionRequest;
+    /// use triton_distributed_llm::protocols::openai::completions::CompletionRequest;
     ///
     /// let request = CompletionRequest::builder()
     ///     .prompt("What is the meaning of life?")
@@ -271,7 +272,7 @@ impl ChatCompletionRequestBuilder {
     /// Builds and validates the ChatCompletionRequest
     ///
     /// ```rust
-    /// use triton_llm::protocols::openai::chat_completions::ChatCompletionRequest;
+    /// use triton_distributed_llm::protocols::openai::chat_completions::ChatCompletionRequest;
     ///
     /// let request = ChatCompletionRequest::builder()
     ///     .model("mixtral-8x7b-instruct-v0.1")