[CI][NPU] Remove model tests from PR-test pipeline (#38011)

.github/workflows/pr-test-npu.yml

@@ -305,32 +305,6 @@ jobs:
       device_type_for_deps: 'a3'
       partitions: '{"size":2,"arr":[0,1]}'
 
-  base-c-test-acc-4-npu-a3:
-    name: base-c-test-acc-4-npu-a3
-    needs: [ check-changes, pr-gate, set-image-config, base-a-test-1-npu-a2, base-b-test-1-npu-a3, base-b-test-2-npu-a3, base-b-test-4-npu-a3, base-b-test-8-npu-a3, base-b-test-16-npu-a3 ]
-    if: ${{ !failure() && !cancelled() && needs.check-changes.outputs.main_package == 'true' }}
-    uses: ./.github/workflows/_npu-single-node-test-stage.yml
-    with:
-      runner: linux-aarch64-a3-800t-4
-      test_type: 'accuracy'
-      test_suite: base-c-test-acc-4-npu-a3
-      image: ${{ needs.set-image-config.outputs.CANN_image_a3 }}
-      install_sglang_deps: true
-      device_type_for_deps: 'a3'
-
-  base-c-test-acc-8-npu-a3:
-    name: base-c-test-acc-8-npu-a3
-    needs: [ check-changes, pr-gate, set-image-config, base-a-test-1-npu-a2, base-b-test-1-npu-a3, base-b-test-2-npu-a3, base-b-test-4-npu-a3, base-b-test-8-npu-a3, base-b-test-16-npu-a3 ]
-    if: ${{ !failure() && !cancelled() && needs.check-changes.outputs.main_package == 'true' }}
-    uses: ./.github/workflows/_npu-single-node-test-stage.yml
-    with:
-      runner: linux-aarch64-a3-800t-8
-      test_type: 'accuracy'
-      test_suite: base-c-test-acc-8-npu-a3
-      image: ${{ needs.set-image-config.outputs.CANN_image_a3 }}
-      install_sglang_deps: true
-      device_type_for_deps: 'a3'
-
   base-c-test-acc-16-npu-a3:
     name: base-c-test-acc-16-npu-a3
     needs: [ check-changes, pr-gate, set-image-config, base-a-test-1-npu-a2, base-b-test-1-npu-a3, base-b-test-2-npu-a3, base-b-test-4-npu-a3, base-b-test-8-npu-a3, base-b-test-16-npu-a3 ]
@@ -346,7 +320,7 @@ jobs:
 
   base-c-test-perf-2-npu-a3:
     name: base-c-test-perf-2-npu-a3
-    needs: [ check-changes, pr-gate, set-image-config, base-a-test-1-npu-a2, base-b-test-1-npu-a3, base-b-test-2-npu-a3, base-b-test-4-npu-a3, base-b-test-8-npu-a3, base-b-test-16-npu-a3, base-c-test-acc-2-npu-a3, base-c-test-acc-4-npu-a3, base-c-test-acc-8-npu-a3, base-c-test-acc-16-npu-a3 ]
+    needs: [ check-changes, pr-gate, set-image-config, base-a-test-1-npu-a2, base-b-test-1-npu-a3, base-b-test-2-npu-a3, base-b-test-4-npu-a3, base-b-test-8-npu-a3, base-b-test-16-npu-a3, base-c-test-acc-2-npu-a3, base-c-test-acc-16-npu-a3 ]
     if: ${{ !failure() && !cancelled() && needs.check-changes.outputs.main_package == 'true' }}
     uses: ./.github/workflows/_npu-single-node-test-stage.yml
     with:
@@ -357,35 +331,9 @@ jobs:
       install_sglang_deps: true
       device_type_for_deps: 'a3'
 
-  base-c-test-perf-4-npu-a3:
-    name: base-c-test-perf-4-npu-a3
-    needs: [ check-changes, pr-gate, set-image-config, base-a-test-1-npu-a2, base-b-test-1-npu-a3, base-b-test-2-npu-a3, base-b-test-4-npu-a3, base-b-test-8-npu-a3, base-b-test-16-npu-a3, base-c-test-acc-2-npu-a3, base-c-test-acc-4-npu-a3, base-c-test-acc-8-npu-a3, base-c-test-acc-16-npu-a3 ]
-    if: ${{ !failure() && !cancelled() && needs.check-changes.outputs.main_package == 'true' }}
-    uses: ./.github/workflows/_npu-single-node-test-stage.yml
-    with:
-      runner: linux-aarch64-a3-800t-4
-      test_type: 'perf'
-      test_suite: base-c-test-perf-4-npu-a3
-      image: ${{ needs.set-image-config.outputs.CANN_image_a3 }}
-      install_sglang_deps: true
-      device_type_for_deps: 'a3'
-
-  base-c-test-perf-8-npu-a3:
-    name: base-c-test-perf-8-npu-a3
-    needs: [ check-changes, pr-gate, set-image-config, base-a-test-1-npu-a2, base-b-test-1-npu-a3, base-b-test-2-npu-a3, base-b-test-4-npu-a3, base-b-test-8-npu-a3, base-b-test-16-npu-a3, base-c-test-acc-2-npu-a3, base-c-test-acc-4-npu-a3, base-c-test-acc-8-npu-a3, base-c-test-acc-16-npu-a3 ]
-    if: ${{ !failure() && !cancelled() && needs.check-changes.outputs.main_package == 'true' }}
-    uses: ./.github/workflows/_npu-single-node-test-stage.yml
-    with:
-      runner: linux-aarch64-a3-800t-8
-      test_type: 'perf'
-      test_suite: base-c-test-perf-8-npu-a3
-      image: ${{ needs.set-image-config.outputs.CANN_image_a3 }}
-      install_sglang_deps: true
-      device_type_for_deps: 'a3'
-
   base-c-test-perf-16-npu-a3:
     name: base-c-test-perf-16-npu-a3
-    needs: [ check-changes, pr-gate, set-image-config, base-a-test-1-npu-a2, base-b-test-1-npu-a3, base-b-test-2-npu-a3, base-b-test-4-npu-a3, base-b-test-8-npu-a3, base-b-test-16-npu-a3, base-c-test-acc-2-npu-a3, base-c-test-acc-4-npu-a3, base-c-test-acc-8-npu-a3, base-c-test-acc-16-npu-a3 ]
+    needs: [ check-changes, pr-gate, set-image-config, base-a-test-1-npu-a2, base-b-test-1-npu-a3, base-b-test-2-npu-a3, base-b-test-4-npu-a3, base-b-test-8-npu-a3, base-b-test-16-npu-a3, base-c-test-acc-2-npu-a3, base-c-test-acc-16-npu-a3 ]
     if: ${{ !failure() && !cancelled() && needs.check-changes.outputs.main_package == 'true' }}
     uses: ./.github/workflows/_npu-single-node-test-stage.yml
     with:
@@ -414,12 +362,8 @@ jobs:
         multimodal-gen-test-2-npu-a3,
 
         base-c-test-acc-2-npu-a3,
-        base-c-test-acc-4-npu-a3,
-        base-c-test-acc-8-npu-a3,
         base-c-test-acc-16-npu-a3,
         base-c-test-perf-2-npu-a3,
-        base-c-test-perf-4-npu-a3,
-        base-c-test-perf-8-npu-a3,
         base-c-test-perf-16-npu-a3,
       ]
     if: always()

test/registered/npu/accuracy/glm4_7_flash/test_npu_glm4_7_flash_1p_gsm8k.py

@@ -6,7 +6,6 @@ from sglang.test.ascend.e2e.test_npu_accuracy_utils import (
 from sglang.test.ascend.e2e.test_npu_performance_utils import GLM_4_7_FLASH_MODEL_PATH
 from sglang.test.ci.ci_register import register_npu_ci
 
-register_npu_ci(est_time=3600, suite="base-c-test-acc-2-npu-a3")
 register_npu_ci(est_time=6500, suite="nightly-acc-2-npu-a3", nightly=True)
 
 ENVS = {

test/registered/npu/accuracy/qwen3_5_9b/test_npu_qwen3_5_9b_bf16_1p_gsm8k.py

@@ -8,7 +8,6 @@ from sglang.test.ascend.e2e.test_npu_performance_utils import (
 )
 from sglang.test.ci.ci_register import register_npu_ci
 
-register_npu_ci(est_time=3600, suite="base-c-test-acc-2-npu-a3")
 register_npu_ci(est_time=2800, suite="nightly-acc-2-npu-a3", nightly=True)
 
 QWEN3_5_9B_ENVS = {

test/registered/npu/accuracy/qwen3_vl_30b_a3b/test_npu_qwen3_vl_30b_a3b_bf16_2p_gsm8k.py

@@ -8,7 +8,6 @@ from sglang.test.ascend.e2e.test_npu_performance_utils import (
 )
 from sglang.test.ci.ci_register import register_npu_ci
 
-register_npu_ci(est_time=3600, suite="base-c-test-acc-4-npu-a3")
 register_npu_ci(est_time=4800, suite="nightly-acc-4-npu-a3", nightly=True)
 
 QWEN3_VL_30B_A3B_ENVS = {

test/registered/npu/accuracy/qwen3_vl_8b/test_npu_qwen3_vl_8b_bf16_2p_gsm8k.py

@@ -8,7 +8,6 @@ from sglang.test.ascend.e2e.test_npu_performance_utils import (
 )
 from sglang.test.ci.ci_register import register_npu_ci
 
-register_npu_ci(est_time=3600, suite="base-c-test-acc-4-npu-a3")
 register_npu_ci(est_time=4800, suite="nightly-acc-4-npu-a3", nightly=True)
 
 QWEN3_VL_8B_ENVS = {

test/registered/npu/accuracy/qwen3_vl_8b_thinking/test_npu_qwen3_vl_8b_thinking_1p_mmmu.py

@@ -9,7 +9,6 @@ from sglang.test.ascend.e2e.test_npu_performance_utils import (
 )
 from sglang.test.ci.ci_register import register_npu_ci
 
-register_npu_ci(est_time=3600, suite="base-c-test-perf-2-npu-a3")
 register_npu_ci(est_time=12000, suite="nightly-acc-2-npu-a3", nightly=True)
 
 _is_pr_pipeline = os.environ.get("GITHUB_EVENT_NAME") == "pull_request"

test/registered/npu/performance/kimi_k2_6/test_npu_kimi_k2_6_w4a8_8p_in3k5_out1k5_20ms.py

@@ -10,7 +10,6 @@ from sglang.test.ascend.e2e.test_npu_performance_utils import (
 )
 from sglang.test.ci.ci_register import register_npu_ci
 
-register_npu_ci(est_time=1200, suite="base-c-test-perf-16-npu-a3")
 register_npu_ci(est_time=1200, suite="nightly-perf-16-npu-a3", nightly=True)
 
 KIMI_K2_6_ENVS = {

test/registered/npu/performance/minimax_m2_5/test_npu_minimax_m2_5_w8a8_4p_in64k_out1k_prefix90_50ms.py

@@ -9,7 +9,6 @@ from sglang.test.ascend.e2e.test_npu_performance_utils import (
 )
 from sglang.test.ci.ci_register import register_npu_ci
 
-register_npu_ci(est_time=3600, suite="base-c-test-perf-8-npu-a3")
 register_npu_ci(est_time=3600, suite="nightly-perf-8-npu-a3", nightly=True)
 
 MINIMAX_M2_5_W8A8_4P_IN64K_OUT1K_PREFIX90_ENVS = {

test/registered/npu/performance/qwen3-8b/test_npu_qwen3_8b_w8a8_1p_in3k5_out1k5_50ms.py

@@ -9,7 +9,6 @@ from sglang.test.ascend.e2e.test_npu_performance_utils import (
 )
 from sglang.test.ci.ci_register import register_npu_ci
 
-register_npu_ci(est_time=3600, suite="base-c-test-perf-2-npu-a3")
 register_npu_ci(est_time=3600, suite="nightly-perf-2-npu-a3", nightly=True)
 
 QWEN3_8B_ENVS = {

test/registered/npu/performance/qwen3_235b_a22b/test_npu_qwen3_235b_w8a8_8p_in3k5_out1k5_50ms.py

@@ -9,7 +9,6 @@ from sglang.test.ascend.e2e.test_npu_performance_utils import (
 )
 from sglang.test.ci.ci_register import register_npu_ci
 
-register_npu_ci(est_time=3600, suite="base-c-test-perf-16-npu-a3")
 register_npu_ci(est_time=3600, suite="nightly-perf-16-npu-a3", nightly=True)
 
 QWEN3_235B_ENVS = {

test/registered/npu/performance/qwen3_30b_a3b/test_npu_qwen3_30b_w8a8_1p_in3k5_out1k5_50ms.py

@@ -9,7 +9,6 @@ from sglang.test.ascend.e2e.test_npu_performance_utils import (
 )
 from sglang.test.ci.ci_register import register_npu_ci
 
-register_npu_ci(est_time=3600, suite="base-c-test-perf-2-npu-a3")
 register_npu_ci(est_time=3600, suite="nightly-perf-2-npu-a3", nightly=True)
 
 QWEN3_30B_A3B_ENVS = {

test/registered/npu/performance/qwen3_32b/test_npu_qwen3_32b_w8a8_2p_in3k5_out1k5_50ms.py

@@ -9,7 +9,6 @@ from sglang.test.ascend.e2e.test_npu_performance_utils import (
 )
 from sglang.test.ci.ci_register import register_npu_ci
 
-register_npu_ci(est_time=3600, suite="base-c-test-perf-4-npu-a3")
 register_npu_ci(est_time=3600, suite="nightly-perf-4-npu-a3", nightly=True)
 
 QWEN3_32B_ENVS = {

test/registered/npu/performance/qwen3_next_80b_a3b_instruct/test_npu_qwen3_next_80b_w8a8_2p_in6k_out1k5_bs16.py

@@ -7,7 +7,6 @@ from sglang.test.ascend.e2e.test_npu_performance_utils import (
 )
 from sglang.test.ci.ci_register import register_npu_ci
 
-register_npu_ci(est_time=3600, suite="base-c-test-perf-4-npu-a3")
 register_npu_ci(est_time=3600, suite="nightly-perf-4-npu-a3", nightly=True)
 
 QWEN3_NEXT_80B_A3B_ENVS = {

test/run_suite.py

@@ -101,12 +101,8 @@ PER_COMMIT_SUITES = {
         "base-b-test-8-npu-a3",
         "base-b-test-16-npu-a3",
         "base-c-test-acc-2-npu-a3",
-        "base-c-test-acc-4-npu-a3",
-        "base-c-test-acc-8-npu-a3",
         "base-c-test-acc-16-npu-a3",
         "base-c-test-perf-2-npu-a3",
-        "base-c-test-perf-4-npu-a3",
-        "base-c-test-perf-8-npu-a3",
         "base-c-test-perf-16-npu-a3",
     ],
     HWBackend.XPU: [