[CI/UT] Fix disaggregated prefill ci (#1313)

### What this PR does / why we need it?
Use eager mode to run disaggregated prefill ci

### Does this PR introduce _any_ user-facing change?
N/A

### How was this patch tested?
CI passed with new existing test.

---------

Signed-off-by: MengqingCao <cmq0113@163.com>
This commit is contained in:
Mengqing Cao
2025-06-24 17:11:00 +08:00
committed by GitHub
parent 9cbce423ce
commit 20767a043c
2 changed files with 7 additions and 1 deletions

View File

@@ -66,6 +66,7 @@ function run_prefill_instance() {
--served-model-name Deepseek \
--max-model-len 2000 \
--trust-remote-code \
--enforce-eager \
--kv-transfer-config "$KV_CONFIG"
}
@@ -119,6 +120,7 @@ function run_decode_instance() {
--max-num-batched-tokens 2000 \
--trust-remote-code \
--gpu-memory-utilization 0.9 \
--enforce-eager \
--kv-transfer-config "$KV_CONFIG"
}