From 7132ae853297047401853730be267cb4e9f40293 Mon Sep 17 00:00:00 2001
From: SILONG ZENG <2609716663@qq.com>
Date: Wed, 10 Dec 2025 14:13:56 +0800
Subject: [PATCH] [CI]Cleanup accurary test (#4861)

### What this PR does / why we need it?
Delete accuracy testing of some models:
- Qwen2-VL-7B-Instruct
- Qwen2.5-VL-7B-Instruct
- gemma-2-9b-it
- DeepSeek-V2-Lite

- vLLM version: v0.12.0
- vLLM main:
https://github.com/vllm-project/vllm/commit/ad32e3e19ccf0526cb6744a5fed09a138a5fb2f9

Signed-off-by: MrZ20 <2609716663@qq.com>
---
 .../workflows/vllm_ascend_test_nightly_a2.yaml   |  3 ---
 tests/e2e/models/configs/DeepSeek-V2-Lite.yaml   | 16 ----------------
 .../e2e/models/configs/Qwen2-VL-7B-Instruct.yaml | 10 ----------
 .../models/configs/Qwen2.5-VL-7B-Instruct.yaml   |  9 ---------
 tests/e2e/models/configs/accuracy.txt            |  4 ----
 tests/e2e/models/configs/gemma-2-9b-it.yaml      | 11 -----------
 6 files changed, 53 deletions(-)
 delete mode 100644 tests/e2e/models/configs/DeepSeek-V2-Lite.yaml
 delete mode 100644 tests/e2e/models/configs/Qwen2-VL-7B-Instruct.yaml
 delete mode 100644 tests/e2e/models/configs/Qwen2.5-VL-7B-Instruct.yaml
 delete mode 100644 tests/e2e/models/configs/gemma-2-9b-it.yaml

diff --git a/.github/workflows/vllm_ascend_test_nightly_a2.yaml b/.github/workflows/vllm_ascend_test_nightly_a2.yaml
index a18cd5d4..d13e79f1 100644
--- a/.github/workflows/vllm_ascend_test_nightly_a2.yaml
+++ b/.github/workflows/vllm_ascend_test_nightly_a2.yaml
@@ -82,7 +82,6 @@ jobs:
           - os: linux-aarch64-a2-1
             model_list:
               - Qwen3-8B
-              - Qwen2.5-VL-7B-Instruct
               - Qwen2-Audio-7B-Instruct
               - Qwen3-8B-W8A8
               - Qwen3-VL-8B-Instruct
@@ -91,7 +90,6 @@ jobs:
           - os: linux-aarch64-a2-1
             model_list:
               - ERNIE-4.5-21B-A3B-PT
-              - gemma-2-9b-it
               - gemma-3-4b-it
               - internlm-7b
               - InternVL3_5-8B-hf
@@ -101,7 +99,6 @@ jobs:
             model_list:
               - Qwen3-30B-A3B
               - Qwen3-VL-30B-A3B-Instruct
-              - DeepSeek-V2-Lite
               - Qwen3-30B-A3B-W8A8
           - os: linux-aarch64-a2-4
             model_list:
diff --git a/tests/e2e/models/configs/DeepSeek-V2-Lite.yaml b/tests/e2e/models/configs/DeepSeek-V2-Lite.yaml
deleted file mode 100644
index c23be35b..00000000
--- a/tests/e2e/models/configs/DeepSeek-V2-Lite.yaml
+++ /dev/null
@@ -1,16 +0,0 @@
-model_name: "deepseek-ai/DeepSeek-V2-Lite"
-hardware: "Atlas A2 Series"
-tasks:
-- name: "gsm8k"
-  metrics:
-  - name: "exact_match,strict-match"
-    value: 0.385
-  - name: "exact_match,flexible-extract"
-    value: 0.385
-tensor_parallel_size: 2
-batch_size: 32
-gpu_memory_utilization: 0.7
-apply_chat_template: False
-fewshot_as_multiturn: False
-trust_remote_code: True
-enforce_eager: False
diff --git a/tests/e2e/models/configs/Qwen2-VL-7B-Instruct.yaml b/tests/e2e/models/configs/Qwen2-VL-7B-Instruct.yaml
deleted file mode 100644
index 576b4de6..00000000
--- a/tests/e2e/models/configs/Qwen2-VL-7B-Instruct.yaml
+++ /dev/null
@@ -1,10 +0,0 @@
-model_name: "Qwen/Qwen2-VL-7B-Instruct"
-hardware: "Atlas A2 Series"
-model: "vllm-vlm"
-tasks:
-- name: "mmmu_val"
-  metrics:
-  - name: "acc,none"
-    value: 0.50
-max_model_len: 8192
-gpu_memory_utilization: 0.7
diff --git a/tests/e2e/models/configs/Qwen2.5-VL-7B-Instruct.yaml b/tests/e2e/models/configs/Qwen2.5-VL-7B-Instruct.yaml
deleted file mode 100644
index 85489899..00000000
--- a/tests/e2e/models/configs/Qwen2.5-VL-7B-Instruct.yaml
+++ /dev/null
@@ -1,9 +0,0 @@
-model_name: "Qwen/Qwen2.5-VL-7B-Instruct"
-hardware: "Atlas A2 Series"
-model: "vllm-vlm"
-tasks:
-- name: "mmmu_val"
-  metrics:
-  - name: "acc,none"
-    value: 0.51
-max_model_len: 8192
diff --git a/tests/e2e/models/configs/accuracy.txt b/tests/e2e/models/configs/accuracy.txt
index c15d7986..b4ab5419 100644
--- a/tests/e2e/models/configs/accuracy.txt
+++ b/tests/e2e/models/configs/accuracy.txt
@@ -1,9 +1,6 @@
-DeepSeek-V2-Lite.yaml
-Qwen2.5-VL-7B-Instruct.yaml
 Qwen3-30B-A3B.yaml
 Qwen3-8B.yaml
 Qwen2-7B.yaml
-Qwen2-VL-7B-Instruct.yaml
 Qwen2-Audio-7B-Instruct.yaml
 Qwen3-VL-30B-A3B-Instruct.yaml
 Qwen3-VL-8B-Instruct.yaml
@@ -11,7 +8,6 @@ Qwen2.5-Omni-7B.yaml
 Meta-Llama-3.1-8B-Instruct.yaml
 InternVL3_5-8B.yaml
 ERNIE-4.5-21B-A3B-PT.yaml
-gemma-2-9b-it.yaml
 gemma-3-4b-it.yaml
 internlm-7b.yaml
 Molmo-7B-D-0924.yaml
diff --git a/tests/e2e/models/configs/gemma-2-9b-it.yaml b/tests/e2e/models/configs/gemma-2-9b-it.yaml
deleted file mode 100644
index 050e2f03..00000000
--- a/tests/e2e/models/configs/gemma-2-9b-it.yaml
+++ /dev/null
@@ -1,11 +0,0 @@
-model_name: "LLM-Research/gemma-2-9b-it"
-hardware: "Atlas A2 Series"
-tasks:
-- name: "gsm8k"
-  metrics:
-  - name: "exact_match,strict-match"
-    value: 0.46
-  - name: "exact_match,flexible-extract"
-    value: 0.79
-num_fewshot: 5
-gpu_memory_utilization: 0.8