chore: merge main into Ling and EXAONE builder migration

Signed-off-by: Chen Cui <chcui@nvidia.com>

Author: cuichenx

.github/actions/test-template/action.yml

@@ -193,14 +193,14 @@ runs:
             export BASH_ENV=/dev/null
           fi
 
-          timeout $(( ${{ inputs.timeout }} * 60 ))s bash tests/${{ inputs.is_unit_test == 'true' && 'unit_tests' || format('functional_tests/launch_scripts/{0}', inputs.script_dir) }}/${{ inputs.script }}.sh || EXIT_CODE=$?
+          timeout $(( ${{ inputs.timeout }} * 60 ))s bash "tests/${{ inputs.is_unit_test == 'true' && 'unit_tests' || format('functional_tests/launch_scripts/{0}', inputs.script_dir) }}/$1.sh" || EXIT_CODE=$?
 
           if [[ "$EXIT_CODE" -eq 124 ]]; then
             echo "::error:: Test timed out after ${{ inputs.timeout }} minutes. Please reach out to the PIC."
           fi
 
           exit $EXIT_CODE
-        '
+        ' bash "$SCRIPT"
 
         RUN_TEST_EOF
         )
@@ -212,9 +212,11 @@ runs:
     - name: Run main script
       id: run-main-script
       shell: bash -e -u -o pipefail {0}
+      env:
+        SCRIPT: ${{ inputs.script }}
       run: |
         echo -e "\033[1;34m┌─ launching test ─────────────────────────────────────────────────────────┐\033[0m"
-        echo -e "\033[1;34m│  script    : ${{ inputs.script }}\033[0m"
+        echo -e "\033[1;34m│  script    : $SCRIPT\033[0m"
         echo -e "\033[1;34m│  runner    : ${{ inputs.runner }}\033[0m"
         echo -e "\033[1;34m│  container : ${{ inputs.container-image }}\033[0m"
         echo -e "\033[1;34m└──────────────────────────────────────────────────────────────────────────┘\033[0m"
@@ -229,6 +231,8 @@ runs:
       id: check
       shell: bash -e -u -o pipefail {0}
       if: always()
+      env:
+        SCRIPT: ${{ inputs.script }}
       run: |
         docker exec nemo_container_${{ github.run_id }} /opt/venv/bin/coverage combine || true
         docker exec nemo_container_${{ github.run_id }} /opt/venv/bin/coverage xml || true
@@ -254,19 +258,19 @@ runs:
           echo -e "\033[1;32m╔══════════════════════════════════════════════════════════════════════════╗\033[0m"
           echo -e "\033[1;32m║                                                                          ║\033[0m"
           echo -e "\033[1;32m║   ✅  PASSED                                                             ║\033[0m"
-          echo -e "\033[1;32m║   ${{ inputs.script }}\033[0m"
+          echo -e "\033[1;32m║   $SCRIPT\033[0m"
           echo -e "\033[1;32m║                                                                          ║\033[0m"
           echo -e "\033[1;32m╚══════════════════════════════════════════════════════════════════════════╝\033[0m"
-          echo "::notice title=Result::✅ ${{ inputs.script }} — PASSED"
+          echo "::notice title=Result::✅ $SCRIPT — PASSED"
           exit 0
         else
           echo -e "\033[1;31m╔══════════════════════════════════════════════════════════════════════════╗\033[0m"
           echo -e "\033[1;31m║                                                                          ║\033[0m"
           echo -e "\033[1;31m║   ❌  FAILED  (exit code: ${EXIT_CODE})                                  ║\033[0m"
-          echo -e "\033[1;31m║   ${{ inputs.script }}\033[0m"
+          echo -e "\033[1;31m║   $SCRIPT\033[0m"
           echo -e "\033[1;31m║                                                                          ║\033[0m"
           echo -e "\033[1;31m╚══════════════════════════════════════════════════════════════════════════╝\033[0m"
-          echo "::error title=Result::❌ ${{ inputs.script }} — FAILED (exit $EXIT_CODE)"
+          echo "::error title=Result::❌ $SCRIPT — FAILED (exit $EXIT_CODE)"
           exit 1
         fi
 

.github/workflows/cicd-main.yml

@@ -729,7 +729,9 @@ jobs:
               name=$(basename "$f" .sh)
               timeout=$(get_timeout "$f")
               runner=$(get_runner "$f")
-              entries="${entries}{\"script\":\"${name}\",\"timeout\":${timeout},\"runner\":\"${runner}\"},"
+              entry=$(jq -cn --arg script "$name" --argjson timeout "$timeout" --arg runner "$runner" \
+                '{script: $script, timeout: $timeout, runner: $runner}')
+              entries="${entries}${entry},"
             done
             matrix="{\"include\":[${entries%,}]}"
             echo "matrix_${tier,,}=${matrix}" | tee -a "$GITHUB_OUTPUT"
@@ -740,7 +742,9 @@ jobs:
             name=$(basename "$f" .sh)
             timeout=$(get_timeout "$f")
             runner=$(get_runner "$f")
-            entries="${entries}{\"script\":\"${name}\",\"timeout\":${timeout},\"runner\":\"${runner}\"},"
+            entry=$(jq -cn --arg script "$name" --argjson timeout "$timeout" --arg runner "$runner" \
+              '{script: $script, timeout: $timeout, runner: $runner}')
+            entries="${entries}${entry},"
           done
           echo "matrix_flaky={\"include\":[${entries%,}]}" | tee -a "$GITHUB_OUTPUT"
 
@@ -978,7 +982,9 @@ jobs:
               name=$(basename "$f" .sh)
               timeout=$(get_timeout "$f")
               runner=$(get_runner "$f")
-              entries="${entries}{\"script\":\"${name}\",\"timeout\":${timeout},\"runner\":\"${runner}\"},"
+              entry=$(jq -cn --arg script "$name" --argjson timeout "$timeout" --arg runner "$runner" \
+                '{script: $script, timeout: $timeout, runner: $runner}')
+              entries="${entries}${entry},"
             done
             echo "matrix_gb200_${tier,,}={\"include\":[${entries%,}]}" | tee -a "$GITHUB_OUTPUT"
           done
@@ -988,7 +994,9 @@ jobs:
             name=$(basename "$f" .sh)
             timeout=$(get_timeout "$f")
             runner=$(get_runner "$f")
-            entries="${entries}{\"script\":\"${name}\",\"timeout\":${timeout},\"runner\":\"${runner}\"},"
+            entry=$(jq -cn --arg script "$name" --argjson timeout "$timeout" --arg runner "$runner" \
+              '{script: $script, timeout: $timeout, runner: $runner}')
+            entries="${entries}${entry},"
           done
           echo "matrix_gb200_flaky={\"include\":[${entries%,}]}" | tee -a "$GITHUB_OUTPUT"
 
@@ -1376,3 +1384,44 @@ jobs:
           path: |
             .coverage
           include-hidden-files: true
+
+  notify-nightly-failure:
+    name: Notify scheduled CI failure
+    runs-on: ubuntu-latest
+    environment: main
+    needs:
+      - ephemeral-runner-routing
+      - pre-flight
+      - configure
+      - lint-check
+      - cicd-wait-in-queue
+      - cicd-compute-build-matrix
+      - cicd-container-build
+      - cicd-import-check
+      - cicd-unit-tests-core
+      - cicd-unit-tests-diffusion
+      - generate-test-matrix
+      - cicd-functional-tests-l0
+      - cicd-functional-tests-l1
+      - cicd-functional-tests-l2
+      - cicd-functional-tests-flaky
+      - generate-gb200-test-matrix
+      - cicd-functional-tests-gb200-l0
+      - cicd-functional-tests-gb200-l1
+      - cicd-functional-tests-gb200-l2
+      - cicd-functional-tests-gb200-flaky
+      - Nemo_CICD_Test
+      - Coverage
+    permissions: {}
+    # Keep setup failures visible even when downstream jobs are skipped.
+    if: >-
+      ${{ always() && !cancelled() && failure() &&
+          github.repository == 'NVIDIA-NeMo/Megatron-Bridge' &&
+          github.event_name == 'schedule' &&
+          github.ref_name == github.event.repository.default_branch }}
+    steps:
+      - name: Notify CI alerts channel
+        uses: NVIDIA-NeMo/FW-CI-templates/.github/actions/notify-ci-failure@631c404d00a9e60afc591cd071d35d2e18f82fc6
+        with:
+          needs-json: ${{ toJSON(needs) }}
+          webhook: ${{ secrets.SLACK_WEBHOOK }}

.github/workflows/link-check.yml

@@ -14,15 +14,15 @@
 
 name: Link check
 
-# Checks external (http/https) links in the docs so broken cross-repo / published
-# links (NVBug 6421302 class) are caught before merge and on a weekly cadence.
+# Checks repository links against the candidate Git tree, then checks external
+# and explicitly versioned HTTP(S) destinations online.
 #
 # Scope note: this job checks http/https URLs only (`--scheme`). Internal
 # cross-page links in the Fern docs use a `.md` extension while the source files
 # are `.mdx`, so filesystem-level checking of relative links would report ~600
-# false positives; that convention is out of scope here. GitHub links are NOT
-# excluded. File links are checked through GitHub's raw-content endpoint to
-# avoid transient 503s from HTML pages; GITHUB_TOKEN supports API fallback.
+# false positives; that convention is out of scope here. Same-repository links
+# to the target branch use the checked tree, including unchanged historical
+# pages. Other file links use raw content to avoid GitHub HTML rendering.
 
 on:
   pull_request:
@@ -40,13 +40,57 @@ jobs:
     steps:
       - name: Checkout
         uses: actions/checkout@11d5960a326750d5838078e36cf38b85af677262 # v4
+        with:
+          # On pull_request this is GitHub's synthetic merge, including forks.
+          # A complete tree needs no parents/history or submodule checkout.
+          ref: ${{ github.sha }}
+          persist-credentials: false
+
+      - name: Record checked revision
+        env:
+          EVENT_SHA: ${{ github.sha }}
+          BASE_SHA: ${{ github.event.pull_request.base.sha }}
+          HEAD_SHA: ${{ github.event.pull_request.head.sha }}
+        run: |
+          test "$(git rev-parse HEAD)" = "$EVENT_SHA"
+          git show --no-patch --format='Checked commit: %H%nTree: %T' HEAD
+          echo "PR base: $BASE_SHA"
+          echo "PR head: $HEAD_SHA"
+
+      - name: Extract all documentation URLs
+        uses: lycheeverse/lychee-action@e7477775783ea5526144ba13e8db5eec57747ce8 # v2
+        with:
+          lycheeVersion: v0.24.2
+          token: ''
+          # Compact avoids the action appending a Markdown report link to the dump.
+          format: compact
+          jobSummary: false
+          output: ${{ runner.temp }}/documentation-links.txt
+          args: >-
+            --dump
+            --scheme https
+            --scheme http
+            --root-dir docs/fern
+            "docs/fern/**/*.mdx"
+            "docs/**/*.md"
 
-      - name: Check links
+      - name: Check repository links in the candidate tree
+        env:
+          REPOSITORY: ${{ github.repository }}
+          TARGET_BRANCH: ${{ github.base_ref || github.ref_name }}
+        run: |
+          python3 scripts/docs/check_repository_links.py \
+            --repository "$REPOSITORY" --branch "$TARGET_BRANCH" \
+            --links-file "$RUNNER_TEMP/documentation-links.txt" \
+            --external-links-file "$RUNNER_TEMP/external-links.txt"
+
+      - name: Check external and versioned links
         uses: lycheeverse/lychee-action@e7477775783ea5526144ba13e8db5eec57747ce8 # v2
         with:
+          lycheeVersion: v0.24.2
           fail: true
-          # TODO(kimi-k3): drop the two kimi_k3 / kimi-k3 excludes below once this PR is
-          # merged. They point at files added by the same PR, so blob/main 404s until then.
+          # An all-local scan is valid; the extraction stage must still be nonempty.
+          failIfEmpty: false
           # Keep requests bounded and check file availability without GitHub HTML rendering.
           args: >-
             --no-progress
@@ -63,9 +107,6 @@ jobs:
             --exclude 'github\.com/NVIDIA-NeMo/Megatron-Bridge/stargazers'
             --exclude 'sched\.com'
             --exclude 'llama\.com'
-            --exclude 'Megatron-Bridge/(blob/)?main/src/megatron/bridge/models/kimi/kimi_k3_'
-            --exclude 'Megatron-Bridge/(blob/)?main/examples/model_verification_cards/kimi-k3/'
-            "docs/fern/**/*.mdx"
-            "docs/**/*.md"
+            "${{ runner.temp }}/external-links.txt"
         env:
           GITHUB_TOKEN: ${{ secrets.GITHUB_TOKEN }}

.main.commit

@@ -1 +1 @@
-a98aed553e25b4e86c59f50be54ac0ea4227b442
+6e7e6ed435e41af04fc41e7007ef862049153fdc

3rdparty/Megatron-LM

@@ -11,9 +11,16 @@
 </div>
 
 ## 📣 News
+
+- [10/16/2026] [**Reflection AI Beam-1.0-501B-A23B**](https://huggingface.co/reflection/Beam-1.0-501B-A23B) is now supported! Megatron Bridge provides HF↔Megatron checkpoint conversion, inference, and BF16 H200 recipes for pretraining, SFT, long-context SFT, and LoRA PEFT. See the [examples README](https://github.com/NVIDIA-NeMo/Megatron-Bridge/blob/main/examples/models/beam/README.md) and [verification card](https://github.com/NVIDIA-NeMo/Megatron-Bridge/blob/main/examples/model_verification_cards/beam/card.yaml) for verified configurations and current limitations.
+
+- [10/15/2026] [**NVIDIA Nemotron 3.5 Super VL**](https://huggingface.co/nvidia/NVIDIA-Nemotron-3.5-Super-VL-120b-a12b-base-BF16) is released! Megatron Bridge supports the 120B-A12B vision-language model with HF↔Megatron checkpoint conversion, image inference, and pretraining, SFT, and LoRA recipes, plus text-only conversion and training workflows. See the [model documentation](docs/models/nemotron/nemotron3.5-super-vl.md) and [verification card](https://github.com/NVIDIA-NeMo/Megatron-Bridge/blob/main/examples/model_verification_cards/nemotron-3.5-super-vl-120b-a12b/card.yaml) for supported workflows, verified configurations, and current limitations.
+
+- [10/15/2026] [**Muse-Spark-Open-816B-A42B**](https://github.com/NVIDIA-NeMo/Megatron-Bridge/blob/main/examples/models/muse_spark_open/README.md) is now supported! Megatron Bridge adds checkpoint conversion, inference, and GB200 training recipes for the 816B-total / 42B-active vision-and-audio-language MoE model: pretraining, full SFT, long-context SFT at 32K and 64K with context parallelism, LoRA, and image-text SFT and LoRA, with HybridEP MoE dispatch. See the [examples README](https://github.com/NVIDIA-NeMo/Megatron-Bridge/blob/main/examples/models/muse_spark_open/README.md) and the [verification card](https://github.com/NVIDIA-NeMo/Megatron-Bridge/blob/main/examples/model_verification_cards/muse-spark-open-816b-a42b/card.yaml) for details.
+
 - [08/19/2026] **Megatron Bridge 0.6.0 released!** Highlights include GLM-5.2 with cuDNN fused DSA for 128K THD-packed context-parallel training, new LLM/VLM/diffusion models, more reliable VLM data and sequence packing, improved checkpoint and LoRA workflows, unified training and conversion launchers, and refreshed performance recipes. Huge thanks to our community contributors! See the [full release notes](https://github.com/NVIDIA-NeMo/Megatron-Bridge/releases/tag/v0.6.0).
 
-- [08/11/2026] **NVIDIA Nemotron 3.5 Lightning is released!** Read the technical blog [here](https://developer.nvidia.com/blog/nvidia-nemotron-3-5-lightning-delivers-fast-accurate-specialized-task-execution-for-long-running-agents/). See the [Nemotron 3.5 Lightning README](https://github.com/NVIDIA-NeMo/Megatron-Bridge/blob/main/examples/models/nemotron/nemotron_3_5_lightning/README.md) for details.
+- [08/11/2026] [**NVIDIA Nemotron 3.5 Lightning**](https://huggingface.co/nvidia/NVIDIA-Nemotron-3.5-Lightning-30B-A3B-BF16) is released! Read the technical blog [here](https://developer.nvidia.com/blog/nvidia-nemotron-3-5-lightning-delivers-fast-accurate-specialized-task-execution-for-long-running-agents/). See the [Nemotron 3.5 Lightning README](https://github.com/NVIDIA-NeMo/Megatron-Bridge/blob/main/examples/models/nemotron/nemotron_3_5_lightning/README.md) for details.
 
 - [07/31/2026] **[K-EXAONE-2](https://huggingface.co/LGAI-EXAONE/K-EXAONE-2.0-750B-A37B) is now supported**! Day-0 support for LG AI Research's K-EXAONE-2 (750B-A37B MoE) is now available on the [k-exaone-2 branch](https://github.com/NVIDIA-NeMo/Megatron-Bridge/tree/k-exaone-2). Users can convert checkpoints, run inference, fine-tune with SFT or PEFT (LoRA), and use the NVFP4 inference-optimized checkpoint released alongside the model. See the branch documentation for setup instructions and examples.
 
@@ -225,6 +232,7 @@ Megatron Bridge provides out-of-the-box bridges and training recipes for a wide
 |----------------|--------------------|
 | [**BAGEL**](docs/models/bagel/bagel.md) | BAGEL-7B-MoT (requires the tested MCore `dev` revision and `bagel` extra) |
 | [**Bailing**](docs/models/bailing/index.md) | Ling 2.0 / Ling MoE V2 (Bailing) |
+| [**Beam**](docs/models/beam/beam.md) | Beam-1.0-501B-A23B |
 | [**BERT**](docs/models/bert/index.md) | Megatron-Style BERT (`MegatronBertForMaskedLM`) |
 | [**DeepSeek**](docs/models/deepseek/index.md) | DeepSeek V3, DeepSeek V4 / V4 Flash |
 | [**Diffusion**](https://github.com/NVIDIA-NeMo/Megatron-Bridge/tree/main/src/megatron/bridge/diffusion/models) | FLUX, LLaDA 1.5, Nemotron-Labs Diffusion, WAN |
@@ -242,6 +250,7 @@ Megatron Bridge provides out-of-the-box bridges and training recipes for a wide
 | [**Xiaomi-MiMo**](docs/models/mimo/index.md) | Xiaomi-MiMo, MiMo-V2-Flash |
 | [**Moonlight**](docs/models/moonlight/index.md) | Moonlight |
 | [**Muse Glimmer**](docs/models/muse_glimmer/muse-glimmer.md) | Muse Glimmer 30B |
+| [**Muse Spark Open**](docs/models/muse_spark_open/muse-spark-open.md) | Muse-Spark-Open-816B-A42B |
 | [**Nemotron**](docs/models/nemotron/index.md) | Nemotron-3 Nano, Nemotron-3 Super, Nemotron-3 Ultra, Nemotron-3 Nano Omni |
 | [**OLMoE**](docs/models/olmoe/index.md) | OLMoE |
 | [**Qwen**](docs/models/qwen/index.md) | Qwen2 / Qwen2.5, Qwen3, Qwen3-MoE, Qwen3 Next, Qwen3.5 (dense/MoE), Qwen2.5-VL, Qwen3-VL, Qwen3.5-VL, Qwen3.6-VL, Qwen2 Audio, Qwen2.5-Omni, Qwen3-Omni, Qwen3-ASR |