Merge branch 'main' into yuya/fix-mimo-terminal-validation

Author: yaoyu-33

.dev.commit

@@ -1 +1 @@
-02fec4b9a193b2f62c7cfaae0d17db282a8fafe6
+4a4de865795f123eda422ebf0a4c13d2418cf3cb

.main.commit

@@ -1 +1 @@
-c9b53d0a87cb926f47115259593ecfeb351ca29f
+f6c33bde4cdc10805ee7157db746d47b758f5f02

3rdparty/Megatron-LM

@@ -235,7 +235,7 @@ Megatron Bridge provides out-of-the-box bridges and training recipes for a wide
 | **Ernie** | [Ernie 4.5 MoE](https://github.com/NVIDIA-NeMo/Megatron-Bridge/tree/main/src/megatron/bridge/models/ernie), [Ernie 4.5 VL MoE](https://github.com/NVIDIA-NeMo/Megatron-Bridge/tree/main/src/megatron/bridge/models/ernie_vl) |
 | [**Falcon**](docs/models/falcon/index.md) | Falcon H1 |
 | [**Gemma**](docs/models/gemma/index.md) | Gemma / Gemma 2 (deprecated), Gemma 3, Gemma 3-VL, Gemma 4 (26B-A4B MoE / 31B dense), Gemma 4-VL (26B-A4B MoE) |
-| [**GLM**](docs/models/glm/index.md) | GLM-4.5 / GLM-4.7 / GLM-4.7-Flash, GLM-4.5V, GLM-5 / GLM-5.1 / GLM-5.2 |
+| [**GLM**](docs/models/glm/index.md) | GLM-4.5 / GLM-4.7 / GLM-4.7-Flash, GLM-4.5V, GLM-5 / GLM-5.1 / GLM-5.2 / [GLM-5.3](docs/models/glm/glm5-2.md) |
 | [**GPT-OSS**](docs/models/gpt_oss/index.md) | GPT-oss |
 | [**HY V3**](https://huggingface.co/tencent/Hy3-preview-Base) | Hy3 preview-Base (HF → Megatron checkpoint conversion) |
 | [**Kimi**](docs/models/kimi/index.md) | Kimi K2, Kimi-K2.5-VL |

README.md

@@ -45,7 +45,7 @@ finally:
 PY
 
 ARG INSTALL_DEEPEP=True
-ARG DEEPEP_COMMIT=34152ae28f80bcc3ee38d7a12cb2ad87cfd4ea72
+ARG DEEPEP_COMMIT=10d4dd7377d5bce900fbb4b80cce863764892b95
 ARG REINSTALL_NVSHMEM=True
 
 ENV PATH="/opt/venv/bin:/usr/local/bin:$PATH" \

docker/Dockerfile.ci

@@ -93,7 +93,7 @@ RUN pip install "fastapi==0.139.1" \
     "jupyterlab>=4.5.10" \
     "jupyter-server>=2.20.0" \
     "mistune>=3.3.0" \
-    "GitPython>=3.1.55" && \
+    "GitPython>=3.1.62,<4" && \
     # Address CVE from transitive dependency
     pip uninstall -y opencv-python-headless nvidia-dali-cuda130
 

docker/Dockerfile.fw_final

@@ -114,6 +114,7 @@ override-dependencies = [
     "nvidia-modelopt; sys_platform == 'never'",          # Installed by MBridge
     "setuptools>=82.0.0",                                # Required by vllm
     "open-clip-torch>=3.2.0",                            # Required by megatron-bridge, override nemo-toolkit's 2.24.0 constraint
+    "hydra-core>=1.3.4,<1.4",                           # CVE-2026-68508
     "lilcom!=1.8.2",
     "cryptography>=48.0.1,<49",                          # Required by megatron-bridge, override nemo-run's constraint
     "opencv-python; sys_platform == 'never'",            # Use the dependency from the base pytorch container

docker/common/fw_pyproject.toml

@@ -1,13 +1,21 @@
 diff --git a/setup.py b/setup.py
-index 63ce332..4e13462 100644
+index d30d25e..941c96d 100644
 --- a/setup.py
 +++ b/setup.py
-@@ -37,7 +37,7 @@ if __name__ == '__main__':
+@@ -52,7 +52,6 @@ def get_extension_hybrid_ep_cpp():
+     # Basic compile arguments
+     compile_args = {
+         "nvcc": [
+-            "-std=c++17",
+             "-Xcompiler",
+             "-fPIC",
+             "--expt-relaxed-constexpr",
+@@ -191,7 +190,7 @@ def get_extension_deep_ep_cpp():
                   '-Wno-sign-compare', '-Wno-reorder', '-Wno-attributes']
      nvcc_flags = ['-O3', '-Xcompiler', '-O3']
      sources = ['csrc/deep_ep.cpp', 'csrc/kernels/runtime.cu', 'csrc/kernels/layout.cu', 'csrc/kernels/intranode.cu']
 -    include_dirs = ['csrc/']
 +    include_dirs = ['csrc/', '/usr/local/cuda/include/cccl/']
      library_dirs = []
      nvcc_dlink = []
-     extra_link_args = []
+     extra_link_args = ['-lcuda']

docker/patches/deepep.patch

@@ -1,6 +1,37 @@
 # GLM-5.2
 
-[GLM-5](https://huggingface.co/zai-org/GLM-5), [GLM-5.1](https://huggingface.co/zai-org/GLM-5.1), and [GLM-5.2](https://huggingface.co/zai-org/GLM-5.2) are large sparse MoE language models with Multi-Latent Attention and Dynamic Sparse Attention. Megatron Bridge supports these checkpoints through the shared `GLM5Bridge`.
+[GLM-5](https://huggingface.co/zai-org/GLM-5), [GLM-5.1](https://huggingface.co/zai-org/GLM-5.1), [GLM-5.2](https://huggingface.co/zai-org/GLM-5.2), and [GLM-5.3](https://huggingface.co/zai-org/GLM-5.3) use the shared `GLM5Bridge` for their MoE, Multi-Latent Attention, and Dynamic Sparse Attention architecture. GLM-5.3 shares GLM-5.2's architecture and import mappings; full-model GLM-5.3 verification remains pending.
+
+
+## GLM-5.2 / GLM-5.3 compatibility
+
+The pinned publisher configs for [GLM-5.2](https://huggingface.co/zai-org/GLM-5.2/blob/cf457fa734ab149ffef225f80893eb38c6ff5cdc/config.json) and [GLM-5.3](https://huggingface.co/zai-org/GLM-5.3/blob/aca966e4e02791568aa6a4ced368624b3d897f42/config.json) have identical architecture fields: 78 decoder layers, 256 routed experts with top-8 routing, MLA, and the same DSA IndexShare pattern. All 59,585 base checkpoint tensor names and shapes also match. AutoBridge resolves both to `GlmMoeDsaForCausalLM` and `GLM5Bridge`; no separate provider is needed. **GLM-5.3-Flash is a different architecture and is outside this support statement.**
+
+The checkpoints have important differences:
+
+- **Precision:** GLM-5.2 stores BF16 weights (plus FP32 router biases). GLM-5.3 stores 59,044 weights as E4M3 FP8 with FP32 scales per 128×128 block. The existing GLM bridge dequantizes these weights to BF16 on import. This is distinct from enabling FP8 training.
+- **Config metadata:** GLM-5.3 adds `quantization_config` and records Transformers 5.15.0 instead of 5.12.0. Use the repository's supported dependency versions.
+- **Tokenizer and generation defaults:** `tokenizer.json`, `tokenizer_config.json`, and `generation_config.json` are identical at those revisions, including token IDs and sampling defaults.
+- **Chat formatting:** GLM-5.3's template adds low reasoning effort, retains historical reasoning by default, always opens a thinking response (it does not honor `enable_thinking=False`), and changes structured tool-result handling and ordering. Load the template from the GLM-5.3 checkpoint; do not substitute GLM-5.2's template.
+
+For config/provider loading without weights:
+
+```python
+from megatron.bridge import AutoBridge
+
+bridge = AutoBridge.from_hf_pretrained(
+    "zai-org/GLM-5.3", revision="aca966e4e02791568aa6a4ced368624b3d897f42"  # pragma: allowlist secret (public HF revision)
+)
+provider = bridge.to_megatron_provider(load_weights=False)
+```
+
+For **BF16 export** using the FP8 source as the reference, explicitly select `--export-weight-dtype bfloat16` with the GPU or distributed-CPU conversion backend (API: `weight_dtype=torch.bfloat16`). This removes FP8 scale tensors from strict source-key validation and writes a BF16 artifact rather than claiming a bitwise FP8 round trip. The serial-CPU backend does not support this option. MTP is disabled by default; the appended MTP layer is outside that default inference graph.
+
+### Verification limits
+
+The recorded commands and results below belong to the checkpoint named in each verification card. GLM-5.2 results do **not** establish GLM-5.3 full-checkpoint conversion/export, forward parity, generation, SFT/PEFT, resume, long-context behavior, or performance. Those GLM-5.3 checks remain unverified. The `glm52_*` recipes are pinned to GLM-5.2 and retain that name and attribution; changing only their display name would not select GLM-5.3 weights or its chat template.
+
+[Production qualification tracker #5476](https://github.com/NVIDIA-NeMo/Megatron-Bridge/issues/5476) remains open. It includes repeated-MTP training semantics, precision controls, dynamic context parallelism, production recipes, checkpoint continuity, convergence, and hardware-specific performance qualification. See also the open [export context-length fix #5996](https://github.com/NVIDIA-NeMo/Megatron-Bridge/pull/5996), [router-bias fix #6036](https://github.com/NVIDIA-NeMo/Megatron-Bridge/pull/6036), and [recipe-selection fix #5608](https://github.com/NVIDIA-NeMo/Megatron-Bridge/pull/5608).
 
 <!-- BEGIN GENERATED VERIFIED CONFIGURATIONS -->
 

docs/fern/versions/nightly/pages/models/glm/glm5-2.mdx

@@ -1,6 +1,37 @@
 # GLM-5
 
-[GLM-5](https://huggingface.co/zai-org/GLM-5), [GLM-5.1](https://huggingface.co/zai-org/GLM-5.1), and [GLM-5.2](https://huggingface.co/zai-org/GLM-5.2) are large sparse MoE language models with Multi-Latent Attention and Dynamic Sparse Attention. Megatron Bridge supports these checkpoints through the shared `GLM5Bridge`.
+[GLM-5](https://huggingface.co/zai-org/GLM-5), [GLM-5.1](https://huggingface.co/zai-org/GLM-5.1), [GLM-5.2](https://huggingface.co/zai-org/GLM-5.2), and [GLM-5.3](https://huggingface.co/zai-org/GLM-5.3) use the shared `GLM5Bridge` for their MoE, Multi-Latent Attention, and Dynamic Sparse Attention architecture. GLM-5.3 shares GLM-5.2's architecture and import mappings; full-model GLM-5.3 verification remains pending.
+
+
+## GLM-5.2 / GLM-5.3 compatibility
+
+The pinned publisher configs for [GLM-5.2](https://huggingface.co/zai-org/GLM-5.2/blob/cf457fa734ab149ffef225f80893eb38c6ff5cdc/config.json) and [GLM-5.3](https://huggingface.co/zai-org/GLM-5.3/blob/aca966e4e02791568aa6a4ced368624b3d897f42/config.json) have identical architecture fields: 78 decoder layers, 256 routed experts with top-8 routing, MLA, and the same DSA IndexShare pattern. All 59,585 base checkpoint tensor names and shapes also match. AutoBridge resolves both to `GlmMoeDsaForCausalLM` and `GLM5Bridge`; no separate provider is needed. **GLM-5.3-Flash is a different architecture and is outside this support statement.**
+
+The checkpoints have important differences:
+
+- **Precision:** GLM-5.2 stores BF16 weights (plus FP32 router biases). GLM-5.3 stores 59,044 weights as E4M3 FP8 with FP32 scales per 128×128 block. The existing GLM bridge dequantizes these weights to BF16 on import. This is distinct from enabling FP8 training.
+- **Config metadata:** GLM-5.3 adds `quantization_config` and records Transformers 5.15.0 instead of 5.12.0. Use the repository's supported dependency versions.
+- **Tokenizer and generation defaults:** `tokenizer.json`, `tokenizer_config.json`, and `generation_config.json` are identical at those revisions, including token IDs and sampling defaults.
+- **Chat formatting:** GLM-5.3's template adds low reasoning effort, retains historical reasoning by default, always opens a thinking response (it does not honor `enable_thinking=False`), and changes structured tool-result handling and ordering. Load the template from the GLM-5.3 checkpoint; do not substitute GLM-5.2's template.
+
+For config/provider loading without weights:
+
+```python
+from megatron.bridge import AutoBridge
+
+bridge = AutoBridge.from_hf_pretrained(
+    "zai-org/GLM-5.3", revision="aca966e4e02791568aa6a4ced368624b3d897f42"  # pragma: allowlist secret (public HF revision)
+)
+provider = bridge.to_megatron_provider(load_weights=False)
+```
+
+For **BF16 export** using the FP8 source as the reference, explicitly select `--export-weight-dtype bfloat16` with the GPU or distributed-CPU conversion backend (API: `weight_dtype=torch.bfloat16`). This removes FP8 scale tensors from strict source-key validation and writes a BF16 artifact rather than claiming a bitwise FP8 round trip. The serial-CPU backend does not support this option. MTP is disabled by default; the appended MTP layer is outside that default inference graph.
+
+### Verification limits
+
+The recorded commands and results below belong to the checkpoint named in each verification card. GLM-5.2 results do **not** establish GLM-5.3 full-checkpoint conversion/export, forward parity, generation, SFT/PEFT, resume, long-context behavior, or performance. Those GLM-5.3 checks remain unverified. The `glm52_*` recipes are pinned to GLM-5.2 and retain that name and attribution; changing only their display name would not select GLM-5.3 weights or its chat template.
+
+[Production qualification tracker #5476](https://github.com/NVIDIA-NeMo/Megatron-Bridge/issues/5476) remains open. It includes repeated-MTP training semantics, precision controls, dynamic context parallelism, production recipes, checkpoint continuity, convergence, and hardware-specific performance qualification. See also the open [export context-length fix #5996](https://github.com/NVIDIA-NeMo/Megatron-Bridge/pull/5996), [router-bias fix #6036](https://github.com/NVIDIA-NeMo/Megatron-Bridge/pull/6036), and [recipe-selection fix #5608](https://github.com/NVIDIA-NeMo/Megatron-Bridge/pull/5608).
 
 <!-- BEGIN GENERATED VERIFIED CONFIGURATIONS -->