BestWishYsh commited on
Commit
0f7e690
·
verified ·
1 Parent(s): 9f0cffe

Update README.md

Browse files
Files changed (1) hide show
  1. README.md +10 -7
README.md CHANGED
@@ -21,17 +21,19 @@ base_model_relation: finetune
21
  <!-- [![arXiv](https://img.shields.io/badge/arXiv-2501.xxxxx-b31b1b.svg?logo=arxiv)](https://arxiv.org/abs/) -->
22
  [![arXiv](https://img.shields.io/badge/Technical--Report-2501.xxxxx-b31b1b.svg?logo=arxiv)](https://github.com/PKU-YuanGroup/Helios-Page/blob/main/helios_technical_report.pdf)
23
  [![Project Page](https://img.shields.io/badge/Project-Website-2ea44f)](https://pku-yuangroup.github.io/Helios-Page)
 
24
  [![HuggingFace](https://img.shields.io/badge/🤗-HuggingFace-blue)](https://huggingface.co/collections/BestWishYsh/helios)
25
  [![ModelScope](https://img.shields.io/badge/🤖-ModelScope-purple)](https://modelscope.cn/collections/BestWishYSH/Helios)
26
 
27
  [![Ascend](https://img.shields.io/badge/Inference-Ascend--NPU-red)](https://www.hiascend.com/)
28
- [![Diffusers](https://img.shields.io/badge/Inference-Diffusers-blueviolet)](https://github.com/huggingface/diffusers)
29
  [![vLLM-Omni](https://img.shields.io/badge/Backend-vLLM--Omni-orange)](https://github.com/vllm-project/vllm-omni/pull/1604)
30
  [![SGLang Diffusion](https://img.shields.io/badge/Backend-SGLang--Diffusion-yellow)](https://github.com/sgl-project/sglang/pull/19782)
31
 
32
 
33
 
34
 
 
35
  </h5>
36
 
37
  <div align="center">
@@ -67,7 +69,7 @@ or you can click <a href="https://github.com/PKU-YuanGroup/Helios-Page/blob/main
67
 
68
  * ⏳⏳⏳ Release the [Technical Report](https://github.com/PKU-YuanGroup/Helios-Page/blob/main/helios_technical_report.pdf) on arXiv.
69
  * `[2025.03.04]` 🚀 Day-0 support for [Ascend-NPU](https://www.hiascend.com),with sincere gratitude to the Ascend Team for their support.
70
- * `[2025.03.04]` 🚀 Day-0 support for [Diffusers](https://github.com/huggingface/diffusers),with special thanks to the HuggingFace Team for their support.
71
  * `[2025.03.04]` 🚀 Day-0 support for [vLLM-Omni](https://github.com/vllm-project/vllm-omni/pull/1604),with heartfelt gratitude to the vLLM Team for their support.
72
  * `[2025.03.04]` 🚀 Day-0 support for [SGLang-Diffusion](https://github.com/sgl-project/sglang/pull/19782),with huge thanks to the SGLang Team for their support.
73
  * `[2025.03.04]` 🔥 We've released the training/inference code and weights of **Helios-Base**, **Helios-Mid** and **Helios-Distilled**.
@@ -270,7 +272,7 @@ For example, let's take Text-to-Video.
270
  --sample-type t2v \
271
  --model ./Helios-Base \
272
  --prompt "A vibrant tropical fish swimming gracefully among colorful coral reefs in a clear, turquoise ocean. The fish has bright blue and yellow scales with a small, distinctive orange spot on its side, its fins moving fluidly. The coral reefs are alive with a variety of marine life, including small schools of colorful fish and sea turtles gliding by. The water is crystal clear, allowing for a view of the sandy ocean floor below. The reef itself is adorned with a mix of hard and soft corals in shades of red, orange, and green. The photo captures the fish from a slightly elevated angle, emphasizing its lively movements and the vivid colors of its surroundings. A close-up shot with dynamic movement." \
273
- --num-frames 600 \
274
  --seed 42 \
275
  --output helios_t2v_base.mp4
276
 
@@ -280,6 +282,7 @@ For example, let's take Text-to-Video.
280
  --prompt "A vibrant tropical fish swimming gracefully among colorful coral reefs in a clear, turquoise ocean. The fish has bright blue and yellow scales with a small, distinctive orange spot on its side, its fins moving fluidly. The coral reefs are alive with a variety of marine life, including small schools of colorful fish and sea turtles gliding by. The water is crystal clear, allowing for a view of the sandy ocean floor below. The reef itself is adorned with a mix of hard and soft corals in shades of red, orange, and green. The photo captures the fish from a slightly elevated angle, emphasizing its lively movements and the vivid colors of its surroundings. A close-up shot with dynamic movement." \
281
  --guidance-scale 5.0 --is-enable-stage2 \
282
  --pyramid-num-inference-steps-list 20 20 20 \
 
283
  --use-cfg-zero-star --use-zero-init --zero-steps 1 \
284
  --output helios_t2v_mid.mp4
285
 
@@ -308,11 +311,11 @@ For example, let's take Helios-Base. **(Native Support)**
308
  ```bash
309
  sglang generate \
310
  --model-path BestWishYsh/Helios-Base \
311
- --prompt "A cat walking on the beach at sunset, cinematic lighting, high quality" \
312
  --negative-prompt "Bright tones, overexposed, static, blurred details, subtitles, style, works, paintings, images, static, overall gray, worst quality, low quality, JPEG compression residue, ugly, incomplete, extra fingers, poorly drawn hands, poorly drawn faces, deformed, disfigured, misshapen limbs, fused fingers, still picture, messy background, three legs, many people in the background, walking backwards" \
313
  --height 384 \
314
  --width 640 \
315
- --num-frames 33 \
316
  --num-inference-steps 50 \
317
  --guidance-scale 5.0
318
  ```
@@ -326,11 +329,11 @@ For example, let's take Helios-Base. **(Diffusers Backend)**
326
  ```bash
327
  sglang generate \
328
  --model-path BestWishYsh/Helios-Base \
329
- --prompt "A cat walking on the beach at sunset, cinematic lighting, high quality" \
330
  --negative-prompt "Bright tones, overexposed, static, blurred details, subtitles, style, works, paintings, images, static, overall gray, worst quality, low quality, JPEG compression residue, ugly, incomplete, extra fingers, poorly drawn hands, poorly drawn faces, deformed, disfigured, misshapen limbs, fused fingers, still picture, messy background, three legs, many people in the background, walking backwards" \
331
  --height 384 \
332
  --width 640 \
333
- --num-frames 33 \
334
  --num-inference-steps 50 \
335
  --guidance-scale 5.0 \
336
  --backend diffusers
 
21
  <!-- [![arXiv](https://img.shields.io/badge/arXiv-2501.xxxxx-b31b1b.svg?logo=arxiv)](https://arxiv.org/abs/) -->
22
  [![arXiv](https://img.shields.io/badge/Technical--Report-2501.xxxxx-b31b1b.svg?logo=arxiv)](https://github.com/PKU-YuanGroup/Helios-Page/blob/main/helios_technical_report.pdf)
23
  [![Project Page](https://img.shields.io/badge/Project-Website-2ea44f)](https://pku-yuangroup.github.io/Helios-Page)
24
+ [![hf_space](https://img.shields.io/badge/🤗-Gradio-00b4d8.svg)](https://huggingface.co/spaces/multimodalart/Helios-Distilled/)
25
  [![HuggingFace](https://img.shields.io/badge/🤗-HuggingFace-blue)](https://huggingface.co/collections/BestWishYsh/helios)
26
  [![ModelScope](https://img.shields.io/badge/🤖-ModelScope-purple)](https://modelscope.cn/collections/BestWishYSH/Helios)
27
 
28
  [![Ascend](https://img.shields.io/badge/Inference-Ascend--NPU-red)](https://www.hiascend.com/)
29
+ [![Diffusers](https://img.shields.io/badge/Inference-Diffusers-blueviolet)](https://github.com/huggingface/diffusers/pull/13208)
30
  [![vLLM-Omni](https://img.shields.io/badge/Backend-vLLM--Omni-orange)](https://github.com/vllm-project/vllm-omni/pull/1604)
31
  [![SGLang Diffusion](https://img.shields.io/badge/Backend-SGLang--Diffusion-yellow)](https://github.com/sgl-project/sglang/pull/19782)
32
 
33
 
34
 
35
 
36
+
37
  </h5>
38
 
39
  <div align="center">
 
69
 
70
  * ⏳⏳⏳ Release the [Technical Report](https://github.com/PKU-YuanGroup/Helios-Page/blob/main/helios_technical_report.pdf) on arXiv.
71
  * `[2025.03.04]` 🚀 Day-0 support for [Ascend-NPU](https://www.hiascend.com),with sincere gratitude to the Ascend Team for their support.
72
+ * `[2025.03.04]` 🚀 Day-0 support for [Diffusers](https://github.com/huggingface/diffusers/pull/13208),with special thanks to the HuggingFace Team for their support.
73
  * `[2025.03.04]` 🚀 Day-0 support for [vLLM-Omni](https://github.com/vllm-project/vllm-omni/pull/1604),with heartfelt gratitude to the vLLM Team for their support.
74
  * `[2025.03.04]` 🚀 Day-0 support for [SGLang-Diffusion](https://github.com/sgl-project/sglang/pull/19782),with huge thanks to the SGLang Team for their support.
75
  * `[2025.03.04]` 🔥 We've released the training/inference code and weights of **Helios-Base**, **Helios-Mid** and **Helios-Distilled**.
 
272
  --sample-type t2v \
273
  --model ./Helios-Base \
274
  --prompt "A vibrant tropical fish swimming gracefully among colorful coral reefs in a clear, turquoise ocean. The fish has bright blue and yellow scales with a small, distinctive orange spot on its side, its fins moving fluidly. The coral reefs are alive with a variety of marine life, including small schools of colorful fish and sea turtles gliding by. The water is crystal clear, allowing for a view of the sandy ocean floor below. The reef itself is adorned with a mix of hard and soft corals in shades of red, orange, and green. The photo captures the fish from a slightly elevated angle, emphasizing its lively movements and the vivid colors of its surroundings. A close-up shot with dynamic movement." \
275
+ --num-frames 99 \
276
  --seed 42 \
277
  --output helios_t2v_base.mp4
278
 
 
282
  --prompt "A vibrant tropical fish swimming gracefully among colorful coral reefs in a clear, turquoise ocean. The fish has bright blue and yellow scales with a small, distinctive orange spot on its side, its fins moving fluidly. The coral reefs are alive with a variety of marine life, including small schools of colorful fish and sea turtles gliding by. The water is crystal clear, allowing for a view of the sandy ocean floor below. The reef itself is adorned with a mix of hard and soft corals in shades of red, orange, and green. The photo captures the fish from a slightly elevated angle, emphasizing its lively movements and the vivid colors of its surroundings. A close-up shot with dynamic movement." \
283
  --guidance-scale 5.0 --is-enable-stage2 \
284
  --pyramid-num-inference-steps-list 20 20 20 \
285
+ --num-frames 99 \
286
  --use-cfg-zero-star --use-zero-init --zero-steps 1 \
287
  --output helios_t2v_mid.mp4
288
 
 
311
  ```bash
312
  sglang generate \
313
  --model-path BestWishYsh/Helios-Base \
314
+ --prompt "A vibrant tropical fish swimming gracefully among colorful coral reefs in a clear, turquoise ocean. The fish has bright blue and yellow scales with a small, distinctive orange spot on its side, its fins moving fluidly. The coral reefs are alive with a variety of marine life, including small schools of colorful fish and sea turtles gliding by. The water is crystal clear, allowing for a view of the sandy ocean floor below. The reef itself is adorned with a mix of hard and soft corals in shades of red, orange, and green. The photo captures the fish from a slightly elevated angle, emphasizing its lively movements and the vivid colors of its surroundings. A close-up shot with dynamic movement." \
315
  --negative-prompt "Bright tones, overexposed, static, blurred details, subtitles, style, works, paintings, images, static, overall gray, worst quality, low quality, JPEG compression residue, ugly, incomplete, extra fingers, poorly drawn hands, poorly drawn faces, deformed, disfigured, misshapen limbs, fused fingers, still picture, messy background, three legs, many people in the background, walking backwards" \
316
  --height 384 \
317
  --width 640 \
318
+ --num-frames 99 \
319
  --num-inference-steps 50 \
320
  --guidance-scale 5.0
321
  ```
 
329
  ```bash
330
  sglang generate \
331
  --model-path BestWishYsh/Helios-Base \
332
+ --prompt "A vibrant tropical fish swimming gracefully among colorful coral reefs in a clear, turquoise ocean. The fish has bright blue and yellow scales with a small, distinctive orange spot on its side, its fins moving fluidly. The coral reefs are alive with a variety of marine life, including small schools of colorful fish and sea turtles gliding by. The water is crystal clear, allowing for a view of the sandy ocean floor below. The reef itself is adorned with a mix of hard and soft corals in shades of red, orange, and green. The photo captures the fish from a slightly elevated angle, emphasizing its lively movements and the vivid colors of its surroundings. A close-up shot with dynamic movement." \
333
  --negative-prompt "Bright tones, overexposed, static, blurred details, subtitles, style, works, paintings, images, static, overall gray, worst quality, low quality, JPEG compression residue, ugly, incomplete, extra fingers, poorly drawn hands, poorly drawn faces, deformed, disfigured, misshapen limbs, fused fingers, still picture, messy background, three legs, many people in the background, walking backwards" \
334
  --height 384 \
335
  --width 640 \
336
+ --num-frames 99 \
337
  --num-inference-steps 50 \
338
  --guidance-scale 5.0 \
339
  --backend diffusers