[AMD] Add Kimi-K2, DeepSeek-V3.2 tests to nightly CI

.github/workflows/nightly-test-amd.yml

@@ -25,18 +25,21 @@ on:
           - 'nightly-perf-2-gpu-text'
           - 'nightly-perf-2-gpu-vlm'
           - 'nightly-accuracy-8-gpu'
-          - 'nightly-accuracy-8-gpu-deepseek-r1'
           # MI30x Accuracy + Performance Tests (combined)
           - 'nightly-8-gpu-grok1-int4'
           - 'nightly-8-gpu-grok2'
           - 'nightly-8-gpu-deepseek-v31'
+          - 'nightly-8-gpu-deepseek-v32'
+          - 'nightly-8-gpu-deepseek-v32-mtp'
+          - 'nightly-8-gpu-kimi-k2'
           # MI35x jobs
           - 'nightly-test-1-gpu-mi35x'
           - 'nightly-accuracy-8-gpu-mi35x'
           - 'nightly-8-gpu-mi35x-grok1-int4'
           - 'nightly-8-gpu-mi35x-grok2'
           - 'nightly-8-gpu-mi35x-deepseek-r1-mxfp4'
           - 'nightly-accuracy-8-gpu-mi35x-deepseek-v32'
+          - 'nightly-accuracy-8-gpu-mi35x-deepseek-v32-mtp'
           - 'nightly-perf-8-gpu-mi35x-deepseek-v32-basic'
           - 'nightly-perf-8-gpu-mi35x-deepseek-v32-mtp'
   workflow_call:
@@ -248,35 +251,6 @@ jobs:
           echo "$(<github_summary.md )" >> $GITHUB_STEP_SUMMARY || true
           exit ${TEST_EXIT_CODE:-0}
 
-  # 8-GPU DeepSeek-R1 Accuracy Test (separate job due to long loading time)
-  nightly-accuracy-8-gpu-deepseek-r1:
-    if: (github.repository == 'sgl-project/sglang' || github.event_name == 'pull_request') && (inputs.job_filter == '' || inputs.job_filter == 'all' || inputs.job_filter == 'nightly-accuracy-8-gpu-deepseek-r1')
-    runs-on: linux-mi325-gpu-8
-    steps:
-      - name: Checkout code
-        uses: actions/checkout@v4
-        with:
-          ref: ${{ inputs.ref || github.ref }}
-
-      - name: Setup docker
-        run: |
-          touch github_summary.md
-          bash scripts/ci/amd/amd_ci_start_container.sh
-        env:
-          GITHUB_WORKSPACE: ${{ github.workspace }}
-
-      - name: Install dependencies
-        run: bash scripts/ci/amd/amd_ci_install_dependency.sh
-
-      - name: Accuracy Test (8-GPU DeepSeek-R1)
-        timeout-minutes: 240
-        run: |
-          bash scripts/ci/amd/amd_ci_exec.sh -w /sglang-checkout/test \
-            -e GITHUB_STEP_SUMMARY="/sglang-checkout/github_summary.md" \
-            python3 run_suite.py --hw amd --suite nightly-amd-accuracy-8-gpu-deepseek-r1 --nightly --timeout-per-file 7200 || TEST_EXIT_CODE=$?
-          echo "$(<github_summary.md )" >> $GITHUB_STEP_SUMMARY || true
-          exit ${TEST_EXIT_CODE:-0}
-
   # ============================================== MI30x Combined Accuracy + Performance Tests ==============================================
   # 8-GPU Grok1-INT4 (Accuracy + Performance combined)
   nightly-8-gpu-grok1-int4:
@@ -407,6 +381,118 @@ jobs:
           echo "$(<github_summary.md )" >> $GITHUB_STEP_SUMMARY || true
           exit ${TEST_EXIT_CODE:-0}
 
+  # 8-GPU DeepSeek-V3.2 (Basic Accuracy + Perf)
+  nightly-8-gpu-deepseek-v32:
+    if: (github.repository == 'sgl-project/sglang' || github.event_name == 'pull_request') && (inputs.job_filter == '' || inputs.job_filter == 'all' || inputs.job_filter == 'nightly-8-gpu-deepseek-v32')
+    runs-on: linux-mi325-gpu-8
+    steps:
+      - name: Checkout code
+        uses: actions/checkout@v4
+        with:
+          ref: ${{ inputs.ref || github.ref }}
+
+      - name: Setup docker
+        run: |
+          touch github_summary.md
+          bash scripts/ci/amd/amd_ci_start_container.sh
+        env:
+          GITHUB_WORKSPACE: ${{ github.workspace }}
+
+      - name: Install dependencies
+        run: bash scripts/ci/amd/amd_ci_install_dependency.sh
+
+      - name: Accuracy Test (8-GPU DeepSeek-V3.2 Basic)
+        timeout-minutes: 120
+        run: |
+          > github_summary.md  # Clear summary file
+          bash scripts/ci/amd/amd_ci_exec.sh -w /sglang-checkout/test \
+            -e GITHUB_STEP_SUMMARY="/sglang-checkout/github_summary.md" \
+            python3 run_suite.py --hw amd --suite nightly-amd-accuracy-8-gpu-deepseek-v32 --nightly --timeout-per-file 3600 || TEST_EXIT_CODE=$?
+          echo "$(<github_summary.md )" >> $GITHUB_STEP_SUMMARY || true
+          exit ${TEST_EXIT_CODE:-0}
+
+      - name: Performance Test (8-GPU DeepSeek-V3.2 Basic)
+        timeout-minutes: 150
+        continue-on-error: true  # Perf test failure doesn't fail the job if accuracy passed
+        run: |
+          > github_summary.md  # Clear summary file
+          bash scripts/ci/amd/amd_ci_exec.sh -w /sglang-checkout/test \
+            -e GITHUB_STEP_SUMMARY="/sglang-checkout/github_summary.md" \
+            python3 run_suite.py --hw amd --suite nightly-perf-8-gpu-deepseek-v32-basic --nightly --timeout-per-file 5400 || TEST_EXIT_CODE=$?
+          echo "$(<github_summary.md )" >> $GITHUB_STEP_SUMMARY || true
+          exit ${TEST_EXIT_CODE:-0}
+
+  # 8-GPU DeepSeek-V3.2 MTP (MTP Accuracy + Perf)
+  nightly-8-gpu-deepseek-v32-mtp:
+    if: (github.repository == 'sgl-project/sglang' || github.event_name == 'pull_request') && (inputs.job_filter == '' || inputs.job_filter == 'all' || inputs.job_filter == 'nightly-8-gpu-deepseek-v32-mtp')
+    runs-on: linux-mi325-gpu-8
+    steps:
+      - name: Checkout code
+        uses: actions/checkout@v4
+        with:
+          ref: ${{ inputs.ref || github.ref }}
+
+      - name: Setup docker
+        run: |
+          touch github_summary.md
+          bash scripts/ci/amd/amd_ci_start_container.sh
+        env:
+          GITHUB_WORKSPACE: ${{ github.workspace }}
+
+      - name: Install dependencies
+        run: bash scripts/ci/amd/amd_ci_install_dependency.sh
+
+      - name: Accuracy Test (8-GPU DeepSeek-V3.2 MTP)
+        timeout-minutes: 120
+        run: |
+          > github_summary.md  # Clear summary file
+          bash scripts/ci/amd/amd_ci_exec.sh -w /sglang-checkout/test \
+            -e GITHUB_STEP_SUMMARY="/sglang-checkout/github_summary.md" \
+            python3 run_suite.py --hw amd --suite nightly-amd-accuracy-8-gpu-deepseek-v32-mtp --nightly --timeout-per-file 3600 || TEST_EXIT_CODE=$?
+          echo "$(<github_summary.md )" >> $GITHUB_STEP_SUMMARY || true
+          exit ${TEST_EXIT_CODE:-0}
+
+      - name: Performance Test (8-GPU DeepSeek-V3.2 MTP)
+        timeout-minutes: 180
+        continue-on-error: true  # Perf test failure doesn't fail the job if accuracy passed
+        run: |
+          > github_summary.md  # Clear summary file
+          bash scripts/ci/amd/amd_ci_exec.sh -w /sglang-checkout/test \
+            -e GITHUB_STEP_SUMMARY="/sglang-checkout/github_summary.md" \
+            python3 run_suite.py --hw amd --suite nightly-perf-8-gpu-deepseek-v32-mtp --nightly --timeout-per-file 7200 || TEST_EXIT_CODE=$?
+          echo "$(<github_summary.md )" >> $GITHUB_STEP_SUMMARY || true
+          exit ${TEST_EXIT_CODE:-0}
+
+  # 8-GPU Kimi-K2 (Accuracy + Speed)
+  nightly-8-gpu-kimi-k2:
+    if: (github.repository == 'sgl-project/sglang' || github.event_name == 'pull_request') && (inputs.job_filter == '' || inputs.job_filter == 'all' || inputs.job_filter == 'nightly-8-gpu-kimi-k2')
+    runs-on: linux-mi325-gpu-8
+    steps:
+      - name: Checkout code
+        uses: actions/checkout@v4
+        with:
+          ref: ${{ inputs.ref || github.ref }}
+
+      - name: Setup docker
+        run: |
+          touch github_summary.md
+          bash scripts/ci/amd/amd_ci_start_container.sh
+        env:
+          GITHUB_WORKSPACE: ${{ github.workspace }}
+
+      - name: Install dependencies
+        run: bash scripts/ci/amd/amd_ci_install_dependency.sh
+
+      - name: Accuracy Test (8-GPU Kimi-K2)
+        timeout-minutes: 120
+        run: |
+          > github_summary.md  # Clear summary file
+          bash scripts/ci/amd/amd_ci_exec.sh -w /sglang-checkout/test \
+            -e GITHUB_STEP_SUMMARY="/sglang-checkout/github_summary.md" \
+            python3 run_suite.py --hw amd --suite nightly-amd-accuracy-8-gpu-kimi-k2 --nightly --timeout-per-file 3600 || TEST_EXIT_CODE=$?
+          echo "$(<github_summary.md )" >> $GITHUB_STEP_SUMMARY || true
+          exit ${TEST_EXIT_CODE:-0}
+
   # ============================================== MI35x Tests ==============================================
   # MI35x 1-GPU tests - platform-agnostic tests that may work on CDNA4 (gfx950)
   nightly-test-1-gpu-mi35x:
@@ -641,6 +727,39 @@ jobs:
           echo "$(<github_summary.md )" >> $GITHUB_STEP_SUMMARY || true
           exit ${TEST_EXIT_CODE:-0}
 
+  # MI35x 8-GPU DeepSeek-V3.2 TP+MTP Accuracy Test
+  nightly-accuracy-8-gpu-mi35x-deepseek-v32-mtp:
+    if: (github.repository == 'sgl-project/sglang' || github.event_name == 'pull_request') && (inputs.job_filter == '' || inputs.job_filter == 'all' || inputs.job_filter == 'nightly-accuracy-8-gpu-mi35x-deepseek-v32-mtp')
+    runs-on: linux-mi35x-gpu-8
+    steps:
+      - name: Checkout code
+        uses: actions/checkout@v4
+        with:
+          ref: ${{ inputs.ref || github.ref }}
+
+      - name: Setup docker
+        run: |
+          touch github_summary.md
+          bash scripts/ci/amd/amd_ci_start_container.sh
+        env:
+          GITHUB_WORKSPACE: ${{ github.workspace }}
+
+      - name: Install dependencies
+        run: |
+          bash scripts/ci/amd/amd_ci_install_dependency.sh
+          # Install tabulate for run_suite.py (missing in MI35x container)
+          bash scripts/ci/amd/amd_ci_exec.sh pip install tabulate
+
+      - name: Accuracy Test MI35x (8-GPU DeepSeek-V3.2 TP+MTP)
+        timeout-minutes: 120
+        run: |
+          > github_summary.md  # Clear summary file
+          bash scripts/ci/amd/amd_ci_exec.sh -w /sglang-checkout/test \
+            -e GITHUB_STEP_SUMMARY="/sglang-checkout/github_summary.md" \
+            python3 run_suite.py --hw amd --suite nightly-amd-accuracy-8-gpu-mi35x-deepseek-v32-mtp --nightly --timeout-per-file 3600 || TEST_EXIT_CODE=$?
+          echo "$(<github_summary.md )" >> $GITHUB_STEP_SUMMARY || true
+          exit ${TEST_EXIT_CODE:-0}
+
   # MI35x 8-GPU DeepSeek-V3.2 Performance Test (Basic)
   nightly-perf-8-gpu-mi35x-deepseek-v32-basic:
     if: (github.repository == 'sgl-project/sglang' || github.event_name == 'pull_request') && (inputs.job_filter == '' || inputs.job_filter == 'all' || inputs.job_filter == 'nightly-perf-8-gpu-mi35x-deepseek-v32-basic')
@@ -698,12 +817,12 @@ jobs:
           bash scripts/ci/amd/amd_ci_exec.sh pip install tabulate
 
       - name: Performance Test MI35x (8-GPU DeepSeek-V3.2 MTP)
-        timeout-minutes: 150
+        timeout-minutes: 180
         run: |
           > github_summary.md  # Clear summary file
           bash scripts/ci/amd/amd_ci_exec.sh -w /sglang-checkout/test \
             -e GITHUB_STEP_SUMMARY="/sglang-checkout/github_summary.md" \
-            python3 run_suite.py --hw amd --suite nightly-perf-8-gpu-mi35x-deepseek-v32-mtp --nightly --timeout-per-file 5400 || TEST_EXIT_CODE=$?
+            python3 run_suite.py --hw amd --suite nightly-perf-8-gpu-mi35x-deepseek-v32-mtp --nightly --timeout-per-file 7200 || TEST_EXIT_CODE=$?
           echo "$(<github_summary.md )" >> $GITHUB_STEP_SUMMARY || true
           exit ${TEST_EXIT_CODE:-0}
 
@@ -719,20 +838,24 @@ jobs:
       - nightly-perf-2-gpu-text
       - nightly-perf-2-gpu-vlm
       - nightly-accuracy-8-gpu
-      - nightly-accuracy-8-gpu-deepseek-r1
       # MI30x Combined Accuracy + Performance Tests
       - nightly-8-gpu-grok1-int4
       - nightly-8-gpu-grok2
       - nightly-8-gpu-deepseek-v31
+      - nightly-8-gpu-deepseek-v32
+      - nightly-8-gpu-deepseek-v32-mtp
+      - nightly-8-gpu-kimi-k2
       # MI35x jobs
       - nightly-test-1-gpu-mi35x
       - nightly-accuracy-8-gpu-mi35x
       - nightly-8-gpu-mi35x-grok1-int4
       - nightly-8-gpu-mi35x-grok2
       - nightly-8-gpu-mi35x-deepseek-r1-mxfp4
       - nightly-accuracy-8-gpu-mi35x-deepseek-v32
-      - nightly-perf-8-gpu-mi35x-deepseek-v32-basic
-      - nightly-perf-8-gpu-mi35x-deepseek-v32-mtp
+      - nightly-accuracy-8-gpu-mi35x-deepseek-v32-mtp
+      # MI35x perf jobs excluded from check - perf failures don't block CI
+      # - nightly-perf-8-gpu-mi35x-deepseek-v32-basic
+      # - nightly-perf-8-gpu-mi35x-deepseek-v32-mtp
     runs-on: ubuntu-latest
     steps:
       - name: Check if any job failed

test/registered/amd/accuracy/mi35x/test_deepseek_r1_eval_mi35x.py

@@ -214,6 +214,9 @@ def test_deepseek_r1_accuracy(self):
                         )
                         passed = acc >= config.accuracy_threshold
                         status = "✅ PASS" if passed else "❌ FAIL"
+                        print(
+                            f"  accuracy={acc:.3f} threshold={config.accuracy_threshold} {status}"
+                        )
 
                         all_results.append(
                             {

test/registered/amd/accuracy/mi35x/test_deepseek_r1_mxfp4_eval_mi35x.py

@@ -239,6 +239,9 @@ def test_deepseek_r1_mxfp4_accuracy(self):
                         )
                         passed = acc >= config.accuracy_threshold
                         status = "✅ PASS" if passed else "❌ FAIL"
+                        print(
+                            f"  accuracy={acc:.3f} threshold={config.accuracy_threshold} {status}"
+                        )
 
                         all_results.append(
                             {