apps/FastDeploy
1
0
Fork 0
mirror of https://github.com/PaddlePaddle/FastDeploy.git synced 2025-10-27 02:20:31 +08:00
Code Issues Actions 2 Packages Projects Releases Wiki Activity

[Executor]CUDAGraph support Speculate Decode (#4258)
Some checks failed
CE Compile Job / ce_job_pre_check (push) Has been cancelled
Details
CE Compile Job / print_ce_job_pre_check_outputs (push) Has been cancelled
Details
CE Compile Job / FD-Clone-Linux (push) Has been cancelled
Details
CE Compile Job / Show Code Archive Output (push) Has been cancelled
Details
CE Compile Job / BUILD_SM8090 (push) Has been cancelled
Details
CE Compile Job / BUILD_SM8689 (push) Has been cancelled
Details
CE Compile Job / CE_UPLOAD (push) Has been cancelled
Details

Browse Source 
* [Executor]CUDAGraph support Speculate Decode

* fix problem

* solve problem

* fix

* fast compile

* CUDAGraph + mtp support eb5(only target model)

* Revert "fast compile"

This reverts commit 3cfe8373ed.

* fix precommit

* solve comment

* fix comment about #pragram unroll

---------

Co-authored-by: gongshaotian <gstain5555@outlook.com>
Co-authored-by: gongshaotian <gstian5555@outlook.com>
This commit is contained in:
Jundong Liu
2025-10-13 15:21:41 +08:00
committed by GitHub
parent 07db281647
commit 0b7a5778ab
16 changed files with 265 additions and 134 deletions
Download Patch File Download Diff File Expand all files Collapse all files

4
fastdeploy/spec_decode/ngram.py
View File

@@ -29,8 +29,8 @@ class NgramProposer(Proposer):
Matching corresponding tokens in input and output as draft tokens.
"""
def __init__(self, cfg: FDConfig):
super().__init__(cfg)
def __init__(self, fd_config: FDConfig):
super().__init__(fd_config)
self.max_ngram_size = self.speculative_config.max_ngram_size
self.input_ids_len = paddle.zeros(shape=[self.max_num_seqs, 1], dtype="int64").cpu()