EngineX/xc-llm-ascend
3
0
Fork 0
Code Issues Pull Requests Projects Releases Wiki Activity

[P/D] check kv extra config and del hccl backend (#4547)

Browse Source 
### What this PR does / why we need it?
check kv extra config & del hccl backend


- vLLM version: v0.12.0
- vLLM main:
ad32e3e19c

---------

Signed-off-by: liziyu <liziyu16@huawei.com>
Co-authored-by: wangxiyuan <wangxiyuan1007@gmail.com>
This commit is contained in:
liziyu
2025-12-07 15:19:42 +08:00
committed by GitHub
parent b91a5f0968
commit 688b1332da
8 changed files with 133 additions and 211 deletions
Download Patch File Download Diff File Expand all files Collapse all files

4
docs/source/tutorials/DeepSeek-V3.1.md
View File

@@ -430,7 +430,6 @@ vllm serve /weights/DeepSeek-V3.1_w8a8mix_mtp \
"engine_id": "0",
"kv_connector_module_path": "vllm_ascend.distributed.mooncake_connector",
"kv_connector_extra_config": {
"use_ascend_direct": true,
"prefill": {
"dp_size": 2,
"tp_size": 8
@@ -510,7 +509,6 @@ vllm serve /weights/DeepSeek-V3.1_w8a8mix_mtp \
"engine_id": "1",
"kv_connector_module_path": "vllm_ascend.distributed.mooncake_connector",
"kv_connector_extra_config": {
"use_ascend_direct": true,
"prefill": {
"dp_size": 2,
"tp_size": 8
@@ -590,7 +588,6 @@ vllm serve /weights/DeepSeek-V3.1_w8a8mix_mtp \
"engine_id": "2",
"kv_connector_module_path": "vllm_ascend.distributed.mooncake_connector",
"kv_connector_extra_config": {
"use_ascend_direct": true,
"prefill": {
"dp_size": 2,
"tp_size": 8
@@ -670,7 +667,6 @@ vllm serve /weights/DeepSeek-V3.1_w8a8mix_mtp \
"engine_id": "3",
"kv_connector_module_path": "vllm_ascend.distributed.mooncake_connector",
"kv_connector_extra_config": {
"use_ascend_direct": true,
"prefill": {
"dp_size": 2,
"tp_size": 8