Hunyuan1.5 (#484)

Thanks to HunyuanVideo Team and ModelTC Team.

---------
Co-authored-by: gushiqiao <975033167@qq.com>
Co-authored-by: gushiqiao <77222802+gushiqiao@users.noreply.github.com>
Co-authored-by: chendingyu <chendingyu1@sensetime.com>
Co-authored-by: XHPlus <xhplus@163.com>
Co-authored-by: wangshankun <wangshankun2011@hotmail.com>
Co-authored-by: STwangyingrui <86730325+STwangyingrui@users.noreply.github.com>
Co-authored-by: root <root@pt-80f094c20fc44a8cad096e5f3dbc962e-worker-0.pt-80f094c20fc44a8cad096e5f3dbc962e.ns-devsft-3460edd0.svc.cluster.local>

scripts/base/base.sh

@@ -14,6 +14,8 @@ export PYTHONPATH=${lightx2v_path}:$PYTHONPATH
 
 # always set false to avoid some warnings
 export TOKENIZERS_PARALLELISM=false
+# set expandable_segments to True to avoid OOM
+export PYTORCH_CUDA_ALLOC_CONF=expandable_segments:True
 
 # =====================================================================================
 # ⚠️  IMPORTANT CONFIGURATION PARAMETERS - READ CAREFULLY AND MODIFY WITH CAUTION ⚠️

scripts/hunyuan_video_15/run_hy15_i2v_480p.sh

+#!/bin/bash
+
+# set path and first
+lightx2v_path=
+model_path=
+
+export CUDA_VISIBLE_DEVICES=0
+
+# set environment variables
+source ${lightx2v_path}/scripts/base/base.sh
+
+python -m lightx2v.infer \
+--seed 123 \
+--model_cls hunyuan_video_1.5 \
+--task i2v \
+--model_path $model_path \
+--config_json ${lightx2v_path}/configs/hunyuan_video_15/hunyuan_video_i2v_480p.json \
+--prompt "Summer beach vacation style, a white cat wearing sunglasses sits on a surfboard. The fluffy-furred feline gazes directly at the camera with a relaxed expression. Blurred beach scenery forms the background featuring crystal-clear waters, distant green hills, and a blue sky dotted with white clouds. The cat assumes a naturally relaxed posture, as if savoring the sea breeze and warm sunlight. A close-up shot highlights the feline's intricate details and the refreshing atmosphere of the seaside." \
+--negative_prompt "" \
+--image_path ${lightx2v_path}/assets/inputs/imgs/img_0.jpg \
+--save_result_path ${lightx2v_path}/save_results/output_lightx2v_hunyuan_video_15_i2v.mp4

scripts/hunyuan_video_15/run_hy15_i2v_720p.sh

+#!/bin/bash
+
+# set path and first
+lightx2v_path=
+model_path=
+
+export CUDA_VISIBLE_DEVICES=0
+
+# set environment variables
+source ${lightx2v_path}/scripts/base/base.sh
+
+python -m lightx2v.infer \
+--seed 123 \
+--model_cls hunyuan_video_1.5 \
+--task i2v \
+--model_path $model_path \
+--config_json ${lightx2v_path}/configs/hunyuan_video_15/hunyuan_video_i2v_720p.json \
+--prompt "Summer beach vacation style, a white cat wearing sunglasses sits on a surfboard. The fluffy-furred feline gazes directly at the camera with a relaxed expression. Blurred beach scenery forms the background featuring crystal-clear waters, distant green hills, and a blue sky dotted with white clouds. The cat assumes a naturally relaxed posture, as if savoring the sea breeze and warm sunlight. A close-up shot highlights the feline's intricate details and the refreshing atmosphere of the seaside." \
+--negative_prompt "" \
+--image_path ${lightx2v_path}/assets/inputs/imgs/img_0.jpg \
+--save_result_path ${lightx2v_path}/save_results/output_lightx2v_hunyuan_video_15_i2v.mp4

scripts/hunyuan_video_15/run_hy15_t2v_480p.sh

+#!/bin/bash
+
+# set path and first
+lightx2v_path=
+model_path=
+
+export CUDA_VISIBLE_DEVICES=2
+
+# set environment variables
+source ${lightx2v_path}/scripts/base/base.sh
+
+python -m lightx2v.infer \
+--seed 123 \
+--model_cls hunyuan_video_1.5 \
+--task t2v \
+--model_path $model_path \
+--config_json ${lightx2v_path}/configs/hunyuan_video_15/hunyuan_video_t2v_480p.json \
+--prompt "A close-up shot captures a scene on a polished, light-colored granite kitchen counter, illuminated by soft natural light from an unseen window. Initially, the frame focuses on a tall, clear glass filled with golden, translucent apple juice standing next to a single, shiny red apple with a green leaf still attached to its stem. The camera moves horizontally to the right. As the shot progresses, a white ceramic plate smoothly enters the frame, revealing a fresh arrangement of about seven or eight more apples, a mix of vibrant reds and greens, piled neatly upon it. A shallow depth of field keeps the focus sharply on the fruit and glass, while the kitchen backsplash in the background remains softly blurred. The scene is in a realistic style." \
+--negative_prompt "" \
+--save_result_path ${lightx2v_path}/save_results/output_lightx2v_hunyuan_video_15_t2v.mp4

scripts/hunyuan_video_15/run_hy15_t2v_720p.sh

+#!/bin/bash
+
+# set path and first
+lightx2v_path=
+model_path=
+
+export CUDA_VISIBLE_DEVICES=2
+
+# set environment variables
+source ${lightx2v_path}/scripts/base/base.sh
+
+python -m lightx2v.infer \
+--seed 123 \
+--model_cls hunyuan_video_1.5 \
+--task t2v \
+--model_path $model_path \
+--config_json ${lightx2v_path}/configs/hunyuan_video_15/hunyuan_video_t2v_720p.json \
+--prompt "A close-up shot captures a scene on a polished, light-colored granite kitchen counter, illuminated by soft natural light from an unseen window. Initially, the frame focuses on a tall, clear glass filled with golden, translucent apple juice standing next to a single, shiny red apple with a green leaf still attached to its stem. The camera moves horizontally to the right. As the shot progresses, a white ceramic plate smoothly enters the frame, revealing a fresh arrangement of about seven or eight more apples, a mix of vibrant reds and greens, piled neatly upon it. A shallow depth of field keeps the focus sharply on the fruit and glass, while the kitchen backsplash in the background remains softly blurred. The scene is in a realistic style." \
+--negative_prompt "" \
+--save_result_path ${lightx2v_path}/save_results/output_lightx2v_hunyuan_video_15_t2v.mp4

scripts/seko_talk/multi_person/01_base.sh

@@ -8,7 +8,6 @@ export CUDA_VISIBLE_DEVICES=0
 # set environment variables
 source ${lightx2v_path}/scripts/base/base.sh
 
-export PYTORCH_CUDA_ALLOC_CONF=expandable_segments:True
 
 export SENSITIVE_LAYER_DTYPE=None
 

scripts/seko_talk/multi_person/03_dist.sh

@@ -8,7 +8,6 @@ export CUDA_VISIBLE_DEVICES=0,1,2,3
 # set environment variables
 source ${lightx2v_path}/scripts/base/base.sh
 
-export PYTORCH_CUDA_ALLOC_CONF=expandable_segments:True
 
 export SENSITIVE_LAYER_DTYPE=None
 

scripts/seko_talk/run_seko_talk_01_base.sh

@@ -8,7 +8,6 @@ export CUDA_VISIBLE_DEVICES=0
 # set environment variables
 source ${lightx2v_path}/scripts/base/base.sh
 
-export PYTORCH_CUDA_ALLOC_CONF=expandable_segments:True
 
 export SENSITIVE_LAYER_DTYPE=None
 

scripts/seko_talk/run_seko_talk_02_fp8.sh

@@ -8,7 +8,6 @@ export CUDA_VISIBLE_DEVICES=0
 # set environment variables
 source ${lightx2v_path}/scripts/base/base.sh
 
-export PYTORCH_CUDA_ALLOC_CONF=expandable_segments:True
 
 export SENSITIVE_LAYER_DTYPE=None
 

scripts/seko_talk/run_seko_talk_03_dist.sh

@@ -8,7 +8,6 @@ export CUDA_VISIBLE_DEVICES=0,1,2,3,4,5,6,7
 # set environment variables
 source ${lightx2v_path}/scripts/base/base.sh
 
-export PYTORCH_CUDA_ALLOC_CONF=expandable_segments:True
 
 export SENSITIVE_LAYER_DTYPE=None
 

scripts/seko_talk/run_seko_talk_04_fp8_dist.sh

@@ -8,7 +8,6 @@ export CUDA_VISIBLE_DEVICES=0,1,2,3,4,5,6,7
 # set environment variables
 source ${lightx2v_path}/scripts/base/base.sh
 
-export PYTORCH_CUDA_ALLOC_CONF=expandable_segments:True
 
 export SENSITIVE_LAYER_DTYPE=None
 

scripts/seko_talk/run_seko_talk_05_offload_fp8_4090.sh

@@ -8,7 +8,6 @@ export CUDA_VISIBLE_DEVICES=0
 # set environment variables
 source ${lightx2v_path}/scripts/base/base.sh
 
-export PYTORCH_CUDA_ALLOC_CONF=expandable_segments:True
 
 export SENSITIVE_LAYER_DTYPE=None
 

scripts/seko_talk/run_seko_talk_06_offload_fp8_H100.sh

@@ -8,7 +8,6 @@ export CUDA_VISIBLE_DEVICES=0
 # set environment variables
 source ${lightx2v_path}/scripts/base/base.sh
 
-export PYTORCH_CUDA_ALLOC_CONF=expandable_segments:True
 
 export SENSITIVE_LAYER_DTYPE=None
 

scripts/seko_talk/run_seko_talk_07_dist_offload.sh

@@ -8,7 +8,6 @@ export CUDA_VISIBLE_DEVICES=0,1,2,3
 # set environment variables
 source ${lightx2v_path}/scripts/base/base.sh
 
-export PYTORCH_CUDA_ALLOC_CONF=expandable_segments:True
 
 export SENSITIVE_LAYER_DTYPE=None
 

scripts/seko_talk/run_seko_talk_08_5B_base.sh

@@ -8,7 +8,6 @@ export CUDA_VISIBLE_DEVICES=0
 # set environment variables
 source ${lightx2v_path}/scripts/base/base.sh
 
-export PYTORCH_CUDA_ALLOC_CONF=expandable_segments:True
 
 export SENSITIVE_LAYER_DTYPE=None
 

scripts/seko_talk/run_seko_talk_09_base_fixed_min_area.sh

@@ -8,7 +8,6 @@ export CUDA_VISIBLE_DEVICES=0
 # set environment variables
 source ${lightx2v_path}/scripts/base/base.sh
 
-export PYTORCH_CUDA_ALLOC_CONF=expandable_segments:True
 
 export SENSITIVE_LAYER_DTYPE=None
 

scripts/seko_talk/run_seko_talk_10_fp8_dist_fixed_min_area.sh

@@ -8,7 +8,6 @@ export CUDA_VISIBLE_DEVICES=0,1,2,3
 # set environment variables
 source ${lightx2v_path}/scripts/base/base.sh
 
-export PYTORCH_CUDA_ALLOC_CONF=expandable_segments:True
 
 export SENSITIVE_LAYER_DTYPE=None
 

scripts/seko_talk/run_seko_talk_11_fp8_dist_fixed_shape.sh

@@ -8,7 +8,6 @@ export CUDA_VISIBLE_DEVICES=0,1,2,3
 # set environment variables
 source ${lightx2v_path}/scripts/base/base.sh
 
-export PYTORCH_CUDA_ALLOC_CONF=expandable_segments:True
 
 export SENSITIVE_LAYER_DTYPE=None
 

scripts/seko_talk/run_seko_talk_12_fp8_dist_fixed_shape_8gpus_1s.sh

@@ -8,7 +8,6 @@ export CUDA_VISIBLE_DEVICES=0,1,2,3,4,5,6,7
 # set environment variables
 source ${lightx2v_path}/scripts/base/base.sh
 
-export PYTORCH_CUDA_ALLOC_CONF=expandable_segments:True
 
 export SENSITIVE_LAYER_DTYPE=None
 

scripts/seko_talk/run_seko_talk_13_fp8_dist_bucket_shape_8gpus_5s_realtime.sh

@@ -8,7 +8,6 @@ export CUDA_VISIBLE_DEVICES=0,1,2,3,4,5,6,7
 # set environment variables
 source ${lightx2v_path}/scripts/base/base.sh
 
-export PYTORCH_CUDA_ALLOC_CONF=expandable_segments:True
 
 export SENSITIVE_LAYER_DTYPE=None