Daniel HiltgenandGitHub 2506f851e4 dflash: add the DFlash2 draft model (#19149) ...

DFlash2 adds block-local dynamic convolutions and a candidate selector
to DFlash. The selector proposes each token conditioned on its
predecessor and returns each position's candidate distribution, which
verification uses to keep the output exact. Like the reference, the
draft samples with the request's temperature only.

A DFlash2 session drafts seven tokens every round on the existing
snapshot verify path instead of following the depth controller, whose
shallower picks lose more accepted tokens than they save in verify time.

qwen3.5-family targets now expose the layer taps, token embeddings and
raw logits a DFlash draft needs.

2026-10-26 16:04:43 -07:00

..

logprob_test.go

mlx, mlxrunner: move the MLX engine out of x/

2026-09-16 14:06:08 -07:00

sample_test.go

dflash: add the DFlash2 draft model (#19149)

2026-10-26 16:04:43 -07:00

sample.go

dflash: add the DFlash2 draft model (#19149)

2026-10-26 16:04:43 -07:00