metrale-speculative
Path: crates/speculative/
Speculative-decoding policy: the MTP gate, the adaptive and DFlash rungs, the n-gram proposer (src/ngram.rs), spec capacity and stats, and scheduler snapshots. See Speculative Decoding (MTP).
Path: crates/speculative/
Speculative-decoding policy: the MTP gate, the adaptive and DFlash rungs, the n-gram proposer (src/ngram.rs), spec capacity and stats, and scheduler snapshots. See Speculative Decoding (MTP).