chore: import upstream snapshot with attribution

examples/megatron/export/full.sh

@@ -0,0 +1,23 @@
+# safetensors -> torch_dist
+CUDA_VISIBLE_DEVICES=0,1,2,3 \
+NPROC_PER_NODE=4 \
+megatron export \
+    --model Qwen/Qwen3-30B-A3B-Instruct-2507 \
+    --output_dir Qwen3-30B-A3B-Instruct-2507-mcore \
+    --to_mcore true \
+    --tensor_model_parallel_size 2 \
+    --expert_model_parallel_size 2 \
+    --pipeline_model_parallel_size 2 \
+    --test_convert_precision true
+
+# torch_dist -> safetensors
+CUDA_VISIBLE_DEVICES=0,1,2,3 \
+NPROC_PER_NODE=4 \
+megatron export \
+    --mcore_model Qwen3-30B-A3B-Instruct-2507-mcore \
+    --output_dir Qwen3-30B-A3B-Instruct-2507-hf \
+    --to_hf true \
+    --tensor_model_parallel_size 2 \
+    --expert_model_parallel_size 2 \
+    --pipeline_model_parallel_size 2 \
+    --test_convert_precision true

examples/megatron/export/lora.sh

@@ -0,0 +1,46 @@
+# torch_dist -> safetensors
+# If you need to perform merge-lora and test precision alignment after merge-lora,
+# simply set `--merge_lora true`
+
+# You can also change `--model safetensors-path` to `--mcore_model torch-dist-path`.
+# These two methods are equivalent, and mcore-bridge will handle it automatically.
+CUDA_VISIBLE_DEVICES=0,1,2,3 \
+NPROC_PER_NODE=4 \
+megatron export \
+    --model Qwen/Qwen3-30B-A3B-Instruct-2507 \
+    --mcore_adapter megatron_output/Qwen3-30B-A3B-Instruct-2507/vx-xxx/checkpoint-xxx \
+    --output_dir megatron_output/Qwen3-30B-A3B-Instruct-2507/vx-xxx/checkpoint-xxx-lora \
+    --merge_lora false \
+    --to_hf true \
+    --tensor_model_parallel_size 2 \
+    --expert_model_parallel_size 2 \
+    --pipeline_model_parallel_size 2 \
+    --test_convert_precision true
+
+# safetensors -> torch_dist
+CUDA_VISIBLE_DEVICES=0,1,2,3 \
+NPROC_PER_NODE=4 \
+megatron export \
+    --model Qwen/Qwen3-30B-A3B-Instruct-2507 \
+    --adapters megatron_output/Qwen3-30B-A3B-Instruct-2507/vx-xxx/checkpoint-xxx-lora \
+    --output_dir megatron_output/Qwen3-30B-A3B-Instruct-2507/vx-xxx/checkpoint-xxx-mcore \
+    --merge_lora false \
+    --to_mcore true \
+    --tensor_model_parallel_size 2 \
+    --expert_model_parallel_size 2 \
+    --pipeline_model_parallel_size 2 \
+    --test_convert_precision true
+
+# Merge-LoRA:
+# torch_dist -> torch_dist
+CUDA_VISIBLE_DEVICES=0,1,2,3 \
+NPROC_PER_NODE=4 \
+megatron export \
+    --model Qwen/Qwen3-30B-A3B-Instruct-2507 \
+    --mcore_adapter megatron_output/Qwen3-30B-A3B-Instruct-2507/vx-xxx/checkpoint-xxx \
+    --output_dir megatron_output/Qwen3-30B-A3B-Instruct-2507/vx-xxx/checkpoint-xxx-merged \
+    --merge_lora true \
+    --to_mcore true \
+    --tensor_model_parallel_size 2 \
+    --expert_model_parallel_size 2 \
+    --pipeline_model_parallel_size 2

examples/megatron/export/merge_lora.sh

@@ -0,0 +1,9 @@
+NPROC_PER_NODE=2 \
+CUDA_VISIBLE_DEVICES=0,1 \
+megatron export \
+    --adapters megatron_output/Qwen2.5-7B-Instruct/vx-xxx/checkpoint-xxx \
+    --tensor_model_parallel_size 2 \
+    --to_hf true \
+    --merge_lora true \
+    --torch_dtype bfloat16 \
+    --output_dir megatron_output/Qwen2.5-7B-Instruct/vx-xxx/checkpoint-xxx-merged