feat: add HunyuanVideo nightly CI test and parameterize diffusion launcher

Add HunyuanVideo-1.5 to the diffusion finetuning CI pipeline alongside
Wan2.1. Parameterize the launcher script to derive model-specific settings
(processor, generate config, model name, frame counts) from the recipe
config name. Also fix a pre-existing T5 layer norm compatibility issue
in finetune.py that affects Hunyuan training with incompatible apex builds.

Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>

Author: pthombre

examples/diffusion/finetune/finetune.py

@@ -16,9 +16,11 @@
 
 from nemo_automodel.components.config._arg_parser import parse_args_and_load_config
 from nemo_automodel.recipes.diffusion.train import TrainDiffusionRecipe
+from nemo_automodel.shared.transformers_patches import patch_t5_layer_norm
 
 
 def main(default_config_path="examples/diffusion/finetune/wan2_1_t2v_flow.yaml"):
+    patch_t5_layer_norm()
     cfg = parse_args_and_load_config(default_config_path)
     recipe = TrainDiffusionRecipe(cfg)
     recipe.setup()

examples/diffusion/finetune/hunyuan_t2v_flow.yaml

@@ -82,3 +82,7 @@ dist_env:
   init_method: "env://"
 
 seed: 42
+
+ci:
+  recipe_owner: pthombre
+  time: "01:30:00"

tests/ci_tests/configs/diffusion_finetune/nightly_recipes.yml

@@ -15,3 +15,4 @@
 examples_dir: diffusion/finetune
 configs:
   - wan2_1_t2v_flow.yaml
+  - hunyuan_t2v_flow.yaml

tests/ci_tests/scripts/diffusion_finetune_launcher.sh

@@ -24,6 +24,32 @@ INFER_DIR="$PIPELINE_DIR/$TEST_NAME/inference_output"
 
 cd /opt/Automodel
 
+# ============================================
+# Derive model-specific settings from config
+# ============================================
+RECIPE_NAME=$(basename "$CONFIG_PATH" .yaml)
+case "$RECIPE_NAME" in
+    wan2_1_t2v_flow*)
+        PROCESSOR="wan"
+        GENERATE_CONFIG="examples/diffusion/generate/configs/generate_wan.yaml"
+        MODEL_NAME="Wan-AI/Wan2.1-T2V-1.3B-Diffusers"
+        INFER_NUM_FRAMES=9
+        PREPROCESS_EXTRA_ARGS=""
+        ;;
+    hunyuan_t2v_flow*)
+        PROCESSOR="hunyuan"
+        GENERATE_CONFIG="examples/diffusion/generate/configs/generate_hunyuan.yaml"
+        MODEL_NAME="hunyuanvideo-community/HunyuanVideo-1.5-Diffusers-720p_t2v"
+        INFER_NUM_FRAMES=5
+        PREPROCESS_EXTRA_ARGS="--target_frames 13"
+        ;;
+    *)
+        echo "ERROR: Unknown recipe '$RECIPE_NAME'. Add a case to diffusion_finetune_launcher.sh."
+        exit 1
+        ;;
+esac
+echo "[config] Recipe=$RECIPE_NAME  Processor=$PROCESSOR  Model=$MODEL_NAME"
+
 # ============================================
 # Stage 1: Download dissolve dataset
 # ============================================
@@ -45,9 +71,10 @@ echo "============================================"
 uv run --extra diffusion python -m tools.diffusion.preprocessing_multiprocess video \
     --video_dir "$DATA_DIR/raw" \
     --output_dir "$DATA_DIR/cache" \
-    --processor wan \
+    --processor "$PROCESSOR" \
     --resolution_preset 512p \
-    --caption_format sidecar
+    --caption_format sidecar \
+    $PREPROCESS_EXTRA_ARGS
 
 # ============================================
 # Stage 3: Finetune
@@ -81,11 +108,11 @@ echo "============================================"
 CKPT_STEP_DIR=$(ls -d $CKPT_DIR/epoch_*_step_* | sort -t_ -k4 -n | tail -1)
 
 uv run --extra diffusion python examples/diffusion/generate/generate.py \
-    --config examples/diffusion/generate/configs/generate_wan.yaml \
-    --model.pretrained_model_name_or_path Wan-AI/Wan2.1-T2V-1.3B-Diffusers \
+    --config "$GENERATE_CONFIG" \
+    --model.pretrained_model_name_or_path "$MODEL_NAME" \
     --model.checkpoint "$CKPT_STEP_DIR" \
     --inference.num_inference_steps 5 \
-    --inference.pipeline_kwargs.num_frames 9 \
+    --inference.pipeline_kwargs.num_frames "$INFER_NUM_FRAMES" \
     --output.output_dir "$INFER_DIR" \
     --vae.enable_slicing true \
     --vae.enable_tiling true