From 4a628f104281d448bb9059cd4eb80f4bb40575fd Mon Sep 17 00:00:00 2001
From: hucong <33891520+underfituu@users.noreply.github.com>
Date: Wed, 8 Apr 2026 21:08:26 +0800
Subject: [PATCH] [UT][v0.18.0] Fix APC nightly UT and TTFT ratio (cherry-pick
 #7468) (#8053)

<!--  Thanks for sending a pull request!

BEFORE SUBMITTING, PLEASE READ
https://docs.vllm.ai/en/latest/contributing/overview.html

-->
### What this PR does / why we need it?
<!--
- Please clarify what changes you are proposing. The purpose of this
section is to outline the changes and how this PR fixes the issue.
If possible, please consider writing useful notes for better and faster
reviews in your PR.

- Please clarify why the changes are needed. For instance, the use case
and bug description.

- Fixes #
-->
Cherry-pick from https://github.com/vllm-project/vllm-ascend/pull/7468

- Fix TTFT ratio threshold from 0.8 to 0.4 for prefix cache benchmarks
- Fix max_out_len values for warm_up and benchmark configs
- Applied to both DeepSeek-R1-0528-W8A8 and Qwen3-32B-Int8 configs

### Does this PR introduce _any_ user-facing change?
<!--
Note that it means *any* user-facing change including all aspects such
as API, interface or other behavior changes.
Documentation-only updates are not considered user-facing changes.
-->

### How was this patch tested?
<!--
CI passed with new added/existing test.
If it was tested in a way different from regular unit tests, please
clarify how you tested step by step, ideally copy and paste-able, so
that other reviewers can test and check, and descendants can verify in
the future.
If tests were not added, please describe why they were not added and/or
why it was difficult to add.
-->

Signed-off-by: underfituu <hzhucong@163.com>
---
 .../configs/Prefix-Cache-DeepSeek-R1-0528-W8A8.yaml       | 8 ++++----
 .../models/configs/Prefix-Cache-Qwen3-32B-Int8.yaml       | 8 ++++----
 2 files changed, 8 insertions(+), 8 deletions(-)

diff --git a/tests/e2e/nightly/single_node/models/configs/Prefix-Cache-DeepSeek-R1-0528-W8A8.yaml b/tests/e2e/nightly/single_node/models/configs/Prefix-Cache-DeepSeek-R1-0528-W8A8.yaml
index ddfbcab6..7629a7ff 100644
--- a/tests/e2e/nightly/single_node/models/configs/Prefix-Cache-DeepSeek-R1-0528-W8A8.yaml
+++ b/tests/e2e/nightly/single_node/models/configs/Prefix-Cache-DeepSeek-R1-0528-W8A8.yaml
@@ -42,7 +42,7 @@ test_cases:
       - metric: "TTFT"
         baseline: "prefix0"
         target: "prefix75"
-        ratio: 0.8
+        ratio: 0.4
         operator: "<"
     benchmarks:
       warm_up:
@@ -51,7 +51,7 @@ test_cases:
         request_conf: vllm_api_stream_chat
         dataset_conf: gsm8k/gsm8k_gen_0_shot_cot_str_perf
         num_prompts: 210
-        max_out_len: 2
+        max_out_len: 1
         batch_size: 1000
         baseline: 0
         threshold: 0.97
@@ -61,7 +61,7 @@ test_cases:
         request_conf: vllm_api_stream_chat
         dataset_conf: gsm8k/gsm8k_gen_0_shot_cot_str_perf
         num_prompts: 210
-        max_out_len: 1500
+        max_out_len: 1
         batch_size: 18
         baseline: 1
         threshold: 0.97
@@ -71,7 +71,7 @@ test_cases:
         request_conf: vllm_api_stream_chat
         dataset_conf: gsm8k/gsm8k_gen_0_shot_cot_str_perf
         num_prompts: 210
-        max_out_len: 1500
+        max_out_len: 1
         batch_size: 18
         baseline: 1
         threshold: 0.97
diff --git a/tests/e2e/nightly/single_node/models/configs/Prefix-Cache-Qwen3-32B-Int8.yaml b/tests/e2e/nightly/single_node/models/configs/Prefix-Cache-Qwen3-32B-Int8.yaml
index 6ead3525..d57368dd 100644
--- a/tests/e2e/nightly/single_node/models/configs/Prefix-Cache-Qwen3-32B-Int8.yaml
+++ b/tests/e2e/nightly/single_node/models/configs/Prefix-Cache-Qwen3-32B-Int8.yaml
@@ -35,7 +35,7 @@ test_cases:
       - metric: "TTFT"
         baseline: "prefix0"
         target: "prefix75"
-        ratio: 0.8
+        ratio: 0.4
         operator: "<"
     benchmarks:
       warm_up:
@@ -44,7 +44,7 @@ test_cases:
         request_conf: vllm_api_stream_chat
         dataset_conf: gsm8k/gsm8k_gen_0_shot_cot_str_perf
         num_prompts: 210
-        max_out_len: 2
+        max_out_len: 1
         batch_size: 1000
         baseline: 0
         threshold: 0.97
@@ -54,7 +54,7 @@ test_cases:
         request_conf: vllm_api_stream_chat
         dataset_conf: gsm8k/gsm8k_gen_0_shot_cot_str_perf
         num_prompts: 210
-        max_out_len: 1500
+        max_out_len: 1
         batch_size: 48
         baseline: 1
         threshold: 0.97
@@ -64,7 +64,7 @@ test_cases:
         request_conf: vllm_api_stream_chat
         dataset_conf: gsm8k/gsm8k_gen_0_shot_cot_str_perf
         num_prompts: 210
-        max_out_len: 1500
+        max_out_len: 1
         batch_size: 48
         baseline: 1
         threshold: 0.97