Skip to content

GitLab

"tests/vscode:/vscode.git/clone" did not exist on "6d89995687e7853946175314a893a11fdc695a0c"
Commit 8f0d4f4d authored by gushiqiao's avatar gushiqiao
Browse files

Update docs

parent 5e74be24
Hide whitespace changes
Inline Side-by-side
Showing with 302 additions and 2 deletions
configs/bench/lightx2v_1.json 0 → 100755
View file @ 8f0d4f4d
{
"infer_steps": 40,
"target_video_length": 81,
"target_height": 480, // 720
"target_width": 832, // 1280
"self_attn_1_type": "sage_attn2",
"cross_attn_1_type": "sage_attn2",
"cross_attn_2_type": "sage_attn2",
"seed": 42, //1234
"sample_guide_scale": 5,
"sample_shift": 5,
"enable_cfg": true,
"cpu_offload": false
}
configs/bench/lightx2v_2.json 0 → 100755
View file @ 8f0d4f4d
{
"infer_steps": 40,
"target_video_length": 81,
"target_height": 480, // 720
"target_width": 832, // 1280
"self_attn_1_type": "sage_attn2",
"cross_attn_1_type": "sage_attn2",
"cross_attn_2_type": "sage_attn2",
"seed": 42, //1234
"sample_guide_scale": 5,
"sample_shift": 5,
"enable_cfg": true,
"cpu_offload": false
}
configs/bench/lightx2v_3.json 0 → 100755
View file @ 8f0d4f4d
{
"infer_steps": 40,
"target_video_length": 81,
"target_height": 480, // 720
"target_width": 832, // 1280
"self_attn_1_type": "sage_attn2",
"cross_attn_1_type": "sage_attn2",
"cross_attn_2_type": "sage_attn2",
"seed": 42, //1234
"sample_guide_scale": 5,
"sample_shift": 5,
"enable_cfg": true,
"cpu_offload": false,
"mm_config": {
"mm_type": "W-fp8-channel-sym-A-fp8-channel-sym-dynamic-Sgl-ActVllm"
}
}
configs/bench/lightx2v_3_distill.json 0 → 100755
View file @ 8f0d4f4d
{
"infer_steps": 40,
"target_video_length": 81,
"target_height": 480, // 720
"target_width": 832, // 1280
"self_attn_1_type": "sage_attn2",
"cross_attn_1_type": "sage_attn2",
"cross_attn_2_type": "sage_attn2",
"seed": 42, //1234
"sample_guide_scale": 5,
"sample_shift": 5,
"enable_cfg": true,
"cpu_offload": false,
"mm_config": {
"mm_type": "W-fp8-channel-sym-A-fp8-channel-sym-dynamic-Sgl-ActVllm"
}
}
configs/bench/lightx2v_4.json 0 → 100755
View file @ 8f0d4f4d
{
"infer_steps": 40,
"target_video_length": 81,
"target_height": 480, // 720
"target_width": 832, // 1280
"self_attn_1_type": "sage_attn2",
"cross_attn_1_type": "sage_attn2",
"cross_attn_2_type": "sage_attn2",
"seed": 42, //1234
"sample_guide_scale": 5,
"sample_shift": 5,
"enable_cfg": true,
"cpu_offload": false,
"mm_config": {
"mm_type": "W-fp8-channel-sym-A-fp8-channel-sym-dynamic-Sgl-ActVllm"
},
"feature_caching": "Tea",
"coefficients": [
[2.57151496e05, -3.54229917e04, 1.40286849e03, -1.35890334e01, 1.32517977e-01],
[-3.02331670e02, 2.23948934e02, -5.25463970e01, 5.87348440e00, -2.01973289e-01]
],
// "coefficients": [
// [8.10705460e03, 2.13393892e03, -3.72934672e02, 1.66203073e01, -4.17769401e-02],
// [-114.36346466, 65.26524496, -18.82220707, 4.91518089, -0.23412683]
// ],
"use_ret_steps": false,
"teacache_thresh": 0.2
}
docs/EN/source/getting_started/benchmark_source.md
View file @ 8f0d4f4d
......@@ -33,7 +33,7 @@
**Test Configuration:**
- **Model**: [Wan2.1-I2V-14B-720P-Lightx2v](https://huggingface.co/lightx2v/Wan2.1-I2V-14B-720P-Lightx2v)
- **Parameters**: infer_steps=40, seed=42, enable_cfg=True
- **Parameters**: infer_steps=40, seed=1234, enable_cfg=True
#### Performance Comparison
......@@ -70,3 +70,6 @@
- **LightX2V_3**: Introduces FP8 quantization technology to significantly reduce computational precision requirements, combined with Tiling VAE technology to optimize memory usage
- **LightX2V_3-Distill**: Based on LightX2V_3 using 4-step distillation model(`infer_step=4`, `enable_cfg=False`), further reducing inference steps while maintaining generation quality.
- **LightX2V_4**: Based on LightX2V_3 with TeaCache(teacache_thresh=0.2) caching reuse technology, achieving acceleration through intelligent redundant computation skipping
- **Configuration Files Reference**: Benchmark-related configuration files and execution scripts are available at:
- [Configuration Files](https://github.com/ModelTC/LightX2V/tree/main/configs/bench) - Contains JSON files with various optimization configurations
- [Execution Scripts](https://github.com/ModelTC/LightX2V/tree/main/scripts/bench) - Contains benchmark execution scripts
docs/ZH_CN/source/getting_started/benchmark_source.md
View file @ 8f0d4f4d
......@@ -62,7 +62,7 @@
---
## 配置说明
## 表格说明
- **Wan2.1 Official**: 基于[Wan2.1官方仓库](https://github.com/Wan-Video/Wan2.1)
- **FastVideo**: 基于[FastVideo官方仓库](https://github.com/hao-ai-lab/FastVideo)，使用SageAttention后端
......@@ -71,3 +71,6 @@
- **LightX2V_3**: 引入FP8量化技术显著减少计算精度要求，结合Tiling VAE技术优化显存使用
- **LightX2V_3-Distill**: 在LightX2V_3基础上使用4步蒸馏模型(`infer_step=4`, `enable_cfg=False`)，进一步减少推理步数并保持生成质量。
- **LightX2V_4**: 在LightX2V_3基础上加入TeaCache(teacache_thresh=0.2)缓存复用技术，通过智能跳过冗余计算实现加速
- **配置文件参考**: 基准测试相关的配置文件和运行脚本可在以下位置获取：
- [配置文件](https://github.com/ModelTC/LightX2V/tree/main/configs/bench) - 包含各种优化配置的JSON文件
- [运行脚本](https://github.com/ModelTC/LightX2V/tree/main/scripts/bench) - 包含基准测试的执行脚本
scripts/bench/run_lightx2v_1.sh 0 → 100755
View file @ 8f0d4f4d
#!/bin/bash
# set path and first
lightx2v_path=/path/to/lightx2v
model_path=/path/to/lightx2v/Wan2.1-I2V-14B-480P-Lightx2v
# model_path=/path/to/lightx2v/Wan2.1-I2V-14B-720P-Lightx2v
# check section
if [ -z "${CUDA_VISIBLE_DEVICES}" ]; then
cuda_devices=0
echo "Warn: CUDA_VISIBLE_DEVICES is not set, using default value: ${cuda_devices}, change at shell script or set env variable."
export CUDA_VISIBLE_DEVICES=${cuda_devices}
fi
if [ -z "${lightx2v_path}" ]; then
echo "Error: lightx2v_path is not set. Please set this variable first."
exit 1
fi
if [ -z "${model_path}" ]; then
echo "Error: model_path is not set. Please set this variable first."
exit 1
fi
export TOKENIZERS_PARALLELISM=false
export PYTHONPATH=${lightx2v_path}:$PYTHONPATH
export ENABLE_PROFILING_DEBUG=true
export ENABLE_GRAPH_MODE=false
python -m lightx2v.infer \
--model_cls wan2.1 \
--task i2v \
--model_path $model_path \
--config_json ${lightx2v_path}/configs/bench/lightx2v_1.json \
--prompt "A close-up cinematic view of a person cooking in a warm,sunlit kitchen, using a wooden spatula to stir-fry a colorful mix of freshvegetables—carrots, broccoli, and bell peppers—in a black frying pan on amodern induction stove. The scene captures the glistening texture of thevegetables, steam gently rising, and subtle reflections on the stove surface.In the background, soft-focus jars, fruits, and a window with natural daylightcreate a cozy atmosphere. The hand motions are smooth and rhythmic, with a realisticsense of motion blur and lighting." \
--negative_prompt "镜头晃动，色调艳丽，过曝，静态，细节模糊不清，字幕，风格，作品，画作，画面，静止，整体发灰，最差质量，低质量，JPEG压缩残留，丑陋的，残缺的，多余的手指，画得不好的手部，画得不好的脸部，畸形的，毁容的，形态畸形的肢体，手指融合，静止不动的画面，杂乱的背景，三条腿，背景人很多，倒着走" \
--image_path ${lightx2v_path}/assets/inputs/imgs/img_2.jpg \
--save_video_path ${lightx2v_path}/save_results/lightx2v_1.mp4
scripts/bench/run_lightx2v_2.sh 0 → 100755
View file @ 8f0d4f4d
#!/bin/bash
# set path and first
lightx2v_path=/path/to/lightx2v
model_path=/path/to/lightx2v/Wan2.1-I2V-14B-480P-Lightx2v
# model_path=/path/to/lightx2v/Wan2.1-I2V-14B-720P-Lightx2v
# check section
if [ -z "${CUDA_VISIBLE_DEVICES}" ]; then
cuda_devices=0
echo "Warn: CUDA_VISIBLE_DEVICES is not set, using default value: ${cuda_devices}, change at shell script or set env variable."
export CUDA_VISIBLE_DEVICES=${cuda_devices}
fi
if [ -z "${lightx2v_path}" ]; then
echo "Error: lightx2v_path is not set. Please set this variable first."
exit 1
fi
if [ -z "${model_path}" ]; then
echo "Error: model_path is not set. Please set this variable first."
exit 1
fi
export TOKENIZERS_PARALLELISM=false
export PYTHONPATH=${lightx2v_path}:$PYTHONPATH
export ENABLE_PROFILING_DEBUG=true
export ENABLE_GRAPH_MODE=false
export DTYPE=BF16
python -m lightx2v.infer \
--model_cls wan2.1 \
--task i2v \
--model_path $model_path \
--config_json ${lightx2v_path}/configs/bench/lightx2v_2.json \
--prompt "A close-up cinematic view of a person cooking in a warm,sunlit kitchen, using a wooden spatula to stir-fry a colorful mix of freshvegetables—carrots, broccoli, and bell peppers—in a black frying pan on amodern induction stove. The scene captures the glistening texture of thevegetables, steam gently rising, and subtle reflections on the stove surface.In the background, soft-focus jars, fruits, and a window with natural daylightcreate a cozy atmosphere. The hand motions are smooth and rhythmic, with a realisticsense of motion blur and lighting." \
--negative_prompt "镜头晃动，色调艳丽，过曝，静态，细节模糊不清，字幕，风格，作品，画作，画面，静止，整体发灰，最差质量，低质量，JPEG压缩残留，丑陋的，残缺的，多余的手指，画得不好的手部，画得不好的脸部，畸形的，毁容的，形态畸形的肢体，手指融合，静止不动的画面，杂乱的背景，三条腿，背景人很多，倒着走" \
--image_path ${lightx2v_path}/assets/inputs/imgs/img_2.jpg \
--save_video_path ${lightx2v_path}/save_results/lightx2v_2.mp4
scripts/bench/run_lightx2v_3.sh 0 → 100755
View file @ 8f0d4f4d
#!/bin/bash
# set path and first
lightx2v_path=/path/to/lightx2v
model_path=/path/to/lightx2v/Wan2.1-I2V-14B-480P-Lightx2v
# model_path=/path/to/lightx2v/Wan2.1-I2V-14B-720P-Lightx2v
# check section
if [ -z "${CUDA_VISIBLE_DEVICES}" ]; then
cuda_devices=0
echo "Warn: CUDA_VISIBLE_DEVICES is not set, using default value: ${cuda_devices}, change at shell script or set env variable."
export CUDA_VISIBLE_DEVICES=${cuda_devices}
fi
if [ -z "${lightx2v_path}" ]; then
echo "Error: lightx2v_path is not set. Please set this variable first."
exit 1
fi
if [ -z "${model_path}" ]; then
echo "Error: model_path is not set. Please set this variable first."
exit 1
fi
export TOKENIZERS_PARALLELISM=false
export PYTHONPATH=${lightx2v_path}:$PYTHONPATH
export ENABLE_PROFILING_DEBUG=true
export ENABLE_GRAPH_MODE=false
export DTYPE=BF16
python -m lightx2v.infer \
--model_cls wan2.1 \
--task i2v \
--model_path $model_path \
--config_json ${lightx2v_path}/configs/bench/lightx2v_3.json \
--prompt "A close-up cinematic view of a person cooking in a warm,sunlit kitchen, using a wooden spatula to stir-fry a colorful mix of freshvegetables—carrots, broccoli, and bell peppers—in a black frying pan on amodern induction stove. The scene captures the glistening texture of thevegetables, steam gently rising, and subtle reflections on the stove surface.In the background, soft-focus jars, fruits, and a window with natural daylightcreate a cozy atmosphere. The hand motions are smooth and rhythmic, with a realisticsense of motion blur and lighting." \
--negative_prompt "镜头晃动，色调艳丽，过曝，静态，细节模糊不清，字幕，风格，作品，画作，画面，静止，整体发灰，最差质量，低质量，JPEG压缩残留，丑陋的，残缺的，多余的手指，画得不好的手部，画得不好的脸部，畸形的，毁容的，形态畸形的肢体，手指融合，静止不动的画面，杂乱的背景，三条腿，背景人很多，倒着走" \
--image_path ${lightx2v_path}/assets/inputs/imgs/img_2.jpg \
--save_video_path ${lightx2v_path}/save_results/lightx2v_3.mp4
scripts/bench/run_lightx2v_3_distill.sh 0 → 100755
View file @ 8f0d4f4d
#!/bin/bash
# set path and first
lightx2v_path=/path/to/lightx2v
model_path=/path/to/lightx2v/Wan2.1-I2V-14B-480P-StepDistill-CfgDistill-Lightx2v
# model_path=/path/to/lightx2v/Wan2.1-I2V-14B-720P-StepDistill-CfgDistill-Lightx2v
# check section
if [ -z "${CUDA_VISIBLE_DEVICES}" ]; then
cuda_devices=0
echo "Warn: CUDA_VISIBLE_DEVICES is not set, using default value: ${cuda_devices}, change at shell script or set env variable."
export CUDA_VISIBLE_DEVICES=${cuda_devices}
fi
if [ -z "${lightx2v_path}" ]; then
echo "Error: lightx2v_path is not set. Please set this variable first."
exit 1
fi
if [ -z "${model_path}" ]; then
echo "Error: model_path is not set. Please set this variable first."
exit 1
fi
export TOKENIZERS_PARALLELISM=false
export PYTHONPATH=${lightx2v_path}:$PYTHONPATH
export ENABLE_PROFILING_DEBUG=true
export ENABLE_GRAPH_MODE=false
export DTYPE=BF16
python -m lightx2v.infer \
--model_cls wan2.1_distill \
--task i2v \
--model_path $model_path \
--config_json ${lightx2v_path}/configs/bench/lightx2v_3_distill.json \
--prompt "A close-up cinematic view of a person cooking in a warm,sunlit kitchen, using a wooden spatula to stir-fry a colorful mix of freshvegetables—carrots, broccoli, and bell peppers—in a black frying pan on amodern induction stove. The scene captures the glistening texture of thevegetables, steam gently rising, and subtle reflections on the stove surface.In the background, soft-focus jars, fruits, and a window with natural daylightcreate a cozy atmosphere. The hand motions are smooth and rhythmic, with a realisticsense of motion blur and lighting." \
--negative_prompt "镜头晃动，色调艳丽，过曝，静态，细节模糊不清，字幕，风格，作品，画作，画面，静止，整体发灰，最差质量，低质量，JPEG压缩残留，丑陋的，残缺的，多余的手指，画得不好的手部，画得不好的脸部，畸形的，毁容的，形态畸形的肢体，手指融合，静止不动的画面，杂乱的背景，三条腿，背景人很多，倒着走" \
--image_path ${lightx2v_path}/assets/inputs/imgs/img_2.jpg \
--save_video_path ${lightx2v_path}/save_results/lightx2v_3_distill.mp4
scripts/bench/run_lightx2v_4.sh 0 → 100755
View file @ 8f0d4f4d
#!/bin/bash
# set path and first
lightx2v_path=/path/to/lightx2v
model_path=/path/to/lightx2v/Wan2.1-I2V-14B-480P-Lightx2v
# model_path=/path/to/lightx2v/Wan2.1-I2V-14B-720P-Lightx2v
# check section
if [ -z "${CUDA_VISIBLE_DEVICES}" ]; then
cuda_devices=0
echo "Warn: CUDA_VISIBLE_DEVICES is not set, using default value: ${cuda_devices}, change at shell script or set env variable."
export CUDA_VISIBLE_DEVICES=${cuda_devices}
fi
if [ -z "${lightx2v_path}" ]; then
echo "Error: lightx2v_path is not set. Please set this variable first."
exit 1
fi
if [ -z "${model_path}" ]; then
echo "Error: model_path is not set. Please set this variable first."
exit 1
fi
export TOKENIZERS_PARALLELISM=false
export PYTHONPATH=${lightx2v_path}:$PYTHONPATH
export ENABLE_PROFILING_DEBUG=true
export ENABLE_GRAPH_MODE=false
export DTYPE=BF16
python -m lightx2v.infer \
--model_cls wan2.1 \
--task i2v \
--model_path $model_path \
--config_json ${lightx2v_path}/configs/bench/lightx2v_4.json \
--prompt "A close-up cinematic view of a person cooking in a warm,sunlit kitchen, using a wooden spatula to stir-fry a colorful mix of freshvegetables—carrots, broccoli, and bell peppers—in a black frying pan on amodern induction stove. The scene captures the glistening texture of thevegetables, steam gently rising, and subtle reflections on the stove surface.In the background, soft-focus jars, fruits, and a window with natural daylightcreate a cozy atmosphere. The hand motions are smooth and rhythmic, with a realisticsense of motion blur and lighting." \
--negative_prompt "镜头晃动，色调艳丽，过曝，静态，细节模糊不清，字幕，风格，作品，画作，画面，静止，整体发灰，最差质量，低质量，JPEG压缩残留，丑陋的，残缺的，多余的手指，画得不好的手部，画得不好的脸部，畸形的，毁容的，形态畸形的肢体，手指融合，静止不动的画面，杂乱的背景，三条腿，背景人很多，倒着走" \
--image_path ${lightx2v_path}/assets/inputs/imgs/img_2.jpg \
--save_video_path ${lightx2v_path}/save_results/lightx2v_4.mp4
Markdown is supported
0% or .
You are about to add 0 people to the discussion. Proceed with caution.
Finish editing this message first!
Please register or to comment