aprender-serve 0.65.1

Pure Rust ML inference engine built from scratch - model serving for GGUF and safetensors
1
2
3
4
5
6
7
8
9
10
11
12
13
14

//! Attention mechanisms: incremental attention, flash decoding, tensor core attention
//!
//! This module implements:
//! - PAR-023: GPU-Resident Incremental Attention
//! - PAR-118: Flash Decoding for parallel KV processing
//! - PAR-065: Tensor Core Attention
//! - Batched attention for multi-sequence processing

use super::*;

include!("gemm_fp16_tensorcore.rs");
include!("incremental_attention_tests.rs");
include!("cb008_decode_wiring_tests.rs");
include!("cb009_kv_handoff_tests.rs");