Expand description
Model executor implementations
Re-exports§
pub use bert_executor::BertModelExecutor;pub use candle_executor::CandleModelExecutor;pub use clip_executor::ClipModelExecutor;pub use qwen2_executor::Qwen2ModelExecutor;pub use qwen3_executor::Qwen3ModelExecutor;pub use stub_executor::StubModelExecutor;
Modules§
- bert_
executor - BERT Model Executor for embeddings
- candle_
executor - Llama model executor using our custom Llama implementation.
- clip_
executor - CLIP Model Executor for multimodal embeddings.
- common
- Common executor utilities — extracted from duplicated code across Qwen3, Qwen2, and Llama executors.
- qwen2_
executor - Qwen2 model executor using Candle
- qwen3_
executor - Qwen3 model executor using Candle
- stub_
executor - Stub model executor for MVP testing and development
- tp_
executor - Tensor-Parallel model executor.