apps/FastDeploy
1
0
Fork 0
mirror of https://github.com/PaddlePaddle/FastDeploy.git synced 2026-04-23 00:17:25 +08:00
Code Issues Actions 19 Packages Projects Releases Wiki Activity

[BugFix] fix async download bug (#5349)

Browse Source 
* fix async download bug

* update log

* Revert "update log"

This reverts commit 5816e602f4.

* update code

* fix mtp bug
This commit is contained in:
kevin
2025-12-05 18:59:12 +08:00
committed by GitHub
parent 5b900667e3
commit c9d7f9e7c3
6 changed files with 59 additions and 22 deletions
Download Patch File Download Diff File Expand all files Collapse all files
fastdeploy/spec_decode/mtp.py
+1 -3
View File
@@ -203,9 +203,7 @@ class MTPProposer(Proposer):
if kv_cache_quant_type == "block_wise_fp8":
kv_cache_scale_shape = [key_cache_shape[0], key_cache_shape[1], key_cache_shape[2]]
local_rank = self.local_rank % self.parallel_config.tensor_parallel_size
if not profile and (
self.cache_config.enable_prefix_caching or self.scheduler_config.splitwise_role != "mixed"
):
if not profile and self.scheduler_config.splitwise_role != "mixed":
cache_kvs_list = []
for i in range(
self.num_main_model_layers,