Module qwen3 

Qwen3 embedding provider via fastembed-rs candle backend.

Uses the candle-nn pure Rust inference engine (no ONNX Runtime). Models are downloaded from HuggingFace on first use.

Supported repos: Qwen/Qwen3-Embedding-0.6B, Qwen/Qwen3-Embedding-8B, Qwen/Qwen3-VL-Embedding-2B (text-only mode).

use llm_kernel::embedding::Qwen3Provider;
use llm_kernel::embedding::EmbeddingProvider;

let provider = Qwen3Provider::new("Qwen/Qwen3-Embedding-0.6B")?;
let result = provider.embed("hello world")?;

Structs

Qwen3Provider

Qwen3 embedding provider backed by candle-nn.

Constants

QWEN3_EMBEDDING_0_6B

Default HuggingFace repo for Qwen3-Embedding-0.6B.

QWEN3_EMBEDDING_8B

Default HuggingFace repo for Qwen3-Embedding-8B.

QWEN3_VL_EMBEDDING_2B

Default HuggingFace repo for Qwen3-VL-Embedding-2B (text-only mode).