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

[CI] Ds32 ep aime2025 (#8496)

Browse Source 
Backport of #7882 to releases/v0.18.0. Adds aime2025 benchmark test for
DeepSeek-V3.2-W8A8 EP with disaggregated prefill on A3 (4-node, 16 NPUs
per node, accuracy benchmark baseline 66.67%).

Signed-off-by: guozr <guozr1997@hotmail.com>
Co-authored-by: guozr <guozr1997@hotmail.com>
This commit is contained in:
starmountain1997
2026-04-21 19:49:06 +08:00
committed by GitHub
parent 862bd8143c
commit 2cb9f76a0f
2 changed files with 222 additions and 0 deletions
Download Patch File Download Diff File Expand all files Collapse all files

3
.github/workflows/schedule_nightly_test_a3.yaml vendored
View File

@@ -119,6 +119,9 @@ jobs:
- name: multi-node-deepseek-v3.2-W8A8-EP
config_file_path: DeepSeek-V3_2-W8A8-EP.yaml
size: 4
- name: multi-node-deepseek-v3.2-W8A8-EP-aime2025
config_file_path: DeepSeek-V3_2-W8A8-EP-aime2025.yaml
size: 4
uses: ./.github/workflows/_e2e_nightly_multi_node.yaml
with:
soc_version: a3