Merge commit 'd803bfe2b1fe7f5e219e50ac20d6801a0a58ac75' as 'vendor/ruvector'

vendor/ruvector/crates/ruvllm/src/models/mod.rs

@@ -0,0 +1,97 @@
+//! Model Architectures for RuvLLM
+//!
+//! This module contains model architecture implementations optimized for
+//! various hardware targets including Apple Neural Engine (ANE), Metal GPU,
+//! and CPU.
+//!
+//! ## Available Models
+//!
+//! | Model | Architecture | Params | ANE Optimized | Use Case |
+//! |-------|--------------|--------|---------------|----------|
+//! | RuvLTRA-Small | Qwen 0.5B | 500M | Yes | Edge inference, mobile |
+//! | RuvLTRA-Medium | Qwen2.5-3B | 3B | Yes | Balanced quality/performance |
+//!
+//! ## Model Selection Guide
+//!
+//! ```text
+//! Model Size vs Performance:
+//!
+//!   RuvLTRA-Small (0.5B)  ████████░░  Good quality, fast inference
+//!                                      ANE: 38 TOPS, ~200 tok/s
+//!
+//!   RuvLTRA-Medium (3B)   ██████████  High quality, moderate speed
+//!                                      GPU/ANE: ~50-80 tok/s, SONA learning
+//!
+//!   Phi-3 (3B)            ██████████  High quality, moderate speed
+//!                                      GPU: Metal, ~50 tok/s
+//!
+//!   Qwen 1.8B             █████████░  Balanced quality/speed
+//!                                      GPU: Metal, ~80 tok/s
+//! ```
+//!
+//! ## Usage
+//!
+//! ### RuvLTRA-Small (0.5B)
+//!
+//! ```rust,ignore
+//! use ruvllm::models::ruvltra::{RuvLtraConfig, RuvLtraModel};
+//!
+//! // Create model with default Qwen 0.5B config
+//! let config = RuvLtraConfig::default();
+//! let model = RuvLtraModel::new(&config)?;
+//!
+//! // Run inference
+//! let logits = model.forward(&input_ids, &positions, None)?;
+//! ```
+//!
+//! ### RuvLTRA-Medium (3B)
+//!
+//! ```rust,ignore
+//! use ruvllm::models::ruvltra_medium::{RuvLtraMediumConfig, RuvLtraMediumModel};
+//!
+//! // Create base variant
+//! let config = RuvLtraMediumConfig::base();
+//! let mut model = RuvLtraMediumModel::new(&config)?;
+//!
+//! // Enable SONA learning hooks at layers 8, 16, 24
+//! model.enable_sona_with_hooks(&[8, 16, 24])?;
+//!
+//! // Run inference with paged attention
+//! let logits = model.forward(&input_ids, &positions)?;
+//! ```
+
+pub mod ruvltra;
+pub mod ruvltra_medium;
+
+// Re-export RuvLTRA-Small types
+pub use ruvltra::{
+    AneDispatcher,
+    AneOptimization,
+    MemoryLayout,
+    QuantizationType,
+    RuvLtraAttention,
+    // Configuration
+    RuvLtraConfig,
+    RuvLtraDecoderLayer,
+    RuvLtraMLP,
+    // Model components
+    RuvLtraModel,
+    // Utilities
+    RuvLtraModelInfo,
+};
+
+// Re-export RuvLTRA-Medium types
+pub use ruvltra_medium::{
+    RuvLtraMediumAttention,
+    // Configuration
+    RuvLtraMediumConfig,
+    RuvLtraMediumDecoderLayer,
+    RuvLtraMediumMLP,
+    // Model components
+    RuvLtraMediumModel,
+    // Utilities
+    RuvLtraMediumModelInfo,
+    RuvLtraMediumQuant,
+    RuvLtraMediumVariant,
+    SonaHookConfig,
+};