Struct Glm5Provider 

pub struct Glm5Provider { /* private fields */ }

GLM-5 FP8 provider targeting a Vast.ai vLLM serverless endpoint.

Implementations

impl Glm5Provider

pub fn new(api_key: String, base_url: String) -> Result<Self>

Create a new Glm5Provider pointing at base_url with api_key.

base_url should be the full base URL including /v1, e.g. https://<vast-endpoint>.vast.ai/v1.

pub fn with_model( api_key: String, base_url: String, model_name: String, ) -> Result<Self>

Create with an explicit model name override.

pub fn normalize_model(model: &str) -> String

Normalize a model string: strip provider prefix (glm5/, glm5:) and map aliases to the canonical model name served by vLLM.

Examples:

• "glm5/glm-5-fp8" → "glm-5-fp8"
• "glm5:glm-5-fp8" → "glm-5-fp8"
• "glm-5-fp8" → "glm-5-fp8"
• "glm-5" → "glm-5-fp8" (alias to fp8 variant)
• "" → DEFAULT_MODEL

Trait Implementations

impl Debug for Glm5Provider

fn fmt(&self, f: &mut Formatter<'_>) -> Result

Formats the value using the given formatter.

impl Provider for Glm5Provider

fn name(&self) -> &str

Provider identifier (e.g. "openai", "bedrock").

fn list_models<'life0, 'async_trait>( &'life0 self, ) -> Pin<Box<dyn Future<Output = Result<Vec<ModelInfo>>> + Send + 'async_trait>>

where Self: 'async_trait, 'life0: 'async_trait,

List models available under this provider.

fn complete<'life0, 'async_trait>( &'life0 self, request: CompletionRequest, ) -> Pin<Box<dyn Future<Output = Result<CompletionResponse>> + Send + 'async_trait>>

where Self: 'async_trait, 'life0: 'async_trait,

Generate a single completion.

fn complete_stream<'life0, 'async_trait>( &'life0 self, request: CompletionRequest, ) -> Pin<Box<dyn Future<Output = Result<BoxStream<'static, StreamChunk>>> + Send + 'async_trait>>

where Self: 'async_trait, 'life0: 'async_trait,

Generate a streaming completion.

fn embed<'life0, 'async_trait>( &'life0 self, _request: EmbeddingRequest, ) -> Pin<Box<dyn Future<Output = Result<EmbeddingResponse>> + Send + 'async_trait>>

where Self: 'async_trait, 'life0: 'async_trait,

Generate embeddings (optional; default returns an error).