1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
//! Embedder foundation for AICX and the Vibecrafted framework.
//!
//! Re-exports the reusable [`aicx_embeddings`] crate so existing consumers
//! keep using `aicx::embedder::*` while `rust-memex` and other workspaces
//! depend on the provider crate directly.
//!
//! Two production backends compile in by default:
//! - **Cloud** (`cloud-embedder` feature) — HTTP POST against an
//! OpenAI-compatible `/v1/embeddings`. Recommended VetCoders production
//! default: zero-install, config-driven URL/model/api_key_env.
//! - **Native GGUF** (`native-embedder` feature) — local llama.cpp
//! inference over an F2LLM/GGUF model resolved from `AICX_EMBEDDER_PATH`
//! or the local HuggingFace cache. Release bundles stay slim and do not
//! silently carry model payloads.
//!
//! Vibecrafted with AI Agents by VetCoders (c)2026 VetCoders
pub use ;
pub use CloudEmbeddingProvider;
pub type EmbedderConfig = EmbeddingConfig;
pub type EmbedderEngine = EmbeddingEngine;
/// Build-time include_bytes embedding is intentionally off for the production
/// GGUF path. Keep this compatibility shim for old diagnostics/tests.
/// GGUF release builds do not expose a static embedded dimension hint because
/// the model is hydrated at install/runtime.