Video wird geladen...

Video konnte nicht geladen werden

Zur Startseite

Today, we open-source Pruna-Qwen-Image-2.1, a set of a few-step LoRA adapters that make Qwen-Image-2.1 by Qwen up to 6.3× faster for image generation and editing. - 𝗛𝗼𝘄 𝗲𝗳𝗳𝗶𝗰𝗶𝗲𝗻𝘁 𝗶𝘀 𝗶𝘁? Generate or edit images in just 5 or 8 steps instead of 40. Choose 5 steps for maximum speed...

PrunaAI's profile picture

Pruna AI

4,554 subscribers

85,115 Aufrufe • vor 2 Tagen •via X (Twitter)

Wissenschaft & Technologie
LIVE

Anya Rossi• Live Now

Private livecam show

17 Kommentare

Profilbild von VOID
VOIDvor 2 Tagen

@Alibaba_Qwen WHAT? WHAT? WHAT? 8 STEPS!!! DAMNNNNNNNNNNNNNNNNNNNNNNNNNNN!!! CC @realrebelai

Profilbild von tobi4s1337
tobi4s1337vor 2 Tagen

@Alibaba_Qwen Any chance you will host these as commercial endpoints on your platform in the future (with a commercial Qwen license)? 👀

Profilbild von Agentsfy
Agentsfyvor 2 Tagen

@Alibaba_Qwen Pruna Qwen Image 2.1 works with ComfyUi: Use 2 strength with 8 steps

Profilbild von Ocean AI
Ocean AIvor 2 Tagen

@huggingface @Alibaba_Qwen amazing🤗

Profilbild von Nid All | AI art and more 🤖
Nid All | AI art and more 🤖vor 2 Tagen

@Alibaba_Qwen @realrebelai the 8-step lora is interesting

Profilbild von NuMust
NuMustvor 2 Tagen

@Alibaba_Qwen the 8-step default with up to three references is useful; i’d want quality scores beside the speedup.

Profilbild von Steven Cheng
Steven Chengvor 2 Tagen

@Alibaba_Qwen 6.3x speedup is wild, but dropping CFG makes me wonder if we're just trading fidelity for raw throughput.

Profilbild von VOID
VOIDvor 2 Tagen

@Alibaba_Qwen What did yall put in this LoRa???? @realrebelai Look what it made you!! 8 Step Edit!! This LoRA turned Qwen Image 2.1 edit and image something else!!

Profilbild von Jai Bharat
Jai Bharatvor 2 Tagen

@Alibaba_Qwen Viggle has already done it. Whats new?

Profilbild von Steven Cheng
Steven Chengvor 2 Tagen

@Alibaba_Qwen Without CFG." That is the real engineering win here.

Profilbild von カイトラ (Kytra) 凯托拉
カイトラ (Kytra) 凯托拉vor 2 Tagen

@Alibaba_Qwen

Profilbild von Evia AI
Evia AIvor 2 Tagen

@Alibaba_Qwen 6.3x faster with just 5 steps is massive - going from 40 to 8 steps while keeping 1K multi-image editing with 3 refs is exactly what production needs. Love that it's just a LoRA drop-in too.

Profilbild von ghosty
ghostyvor 2 Tagen

@Alibaba_Qwen Qwen-Image-Turbo before GTA VI?

Profilbild von IamAlam
IamAlamvor 2 Tagen

@Alibaba_Qwen Let’s connect over DM

Profilbild von unfair.so intern
unfair.so internvor 2 Tagen

@Alibaba_Qwen Faster image models are great. Agents also need fast access to good data. TinyFish's alliance:

Profilbild von karai
karaivor 2 Tagen

@Alibaba_Qwen Make it work on a phone

Profilbild von Cris Lenta
Cris Lentavor 2 Tagen

@Alibaba_Qwen you guys move so fast holy!

Ähnliche Videos

Today, we open-source PrunaVAED, a faster drop-in replacement decoder for video generation with LTX-2.3! - 𝗛𝗼𝘄 𝗲𝗳𝗳𝗶𝗰𝗶𝗲𝗻𝘁 𝗶𝘀 𝗶𝘁? ~1.7-2.1x faster decoder & ~50% lower peak VRAM - 𝗛𝗼𝘄 𝗴𝗼𝗼𝗱 𝗶𝘀 𝗶𝘁? Near-original visual quality & no changes to the latent encoding - 𝗛𝗼𝘄 𝗱𝗼𝗲𝘀 𝗶𝘁 𝘄𝗼𝗿𝗸? We pruned, trained & distilled video decoders to maximize efficiency of the LTX-2.3 series from Lightricks Want to try the fast open-source integration? Check it here on Diffusers: Want to try the fastest video generation endpoint? Check p-video here on API:
0:21

Today, we open-source PrunaVAED, a faster drop-in replacement decoder for video generation with LTX-2.3! - 𝗛𝗼𝘄 𝗲𝗳𝗳𝗶𝗰𝗶𝗲𝗻𝘁 𝗶𝘀 𝗶𝘁? ~1.7-2.1x faster decoder & ~50% lower peak VRAM - 𝗛𝗼𝘄 𝗴𝗼𝗼𝗱 𝗶𝘀 𝗶𝘁? Near-original visual quality & no changes to the latent encoding - 𝗛𝗼𝘄 𝗱𝗼𝗲𝘀 𝗶𝘁 𝘄𝗼𝗿𝗸? We pruned, trained & distilled video decoders to maximize efficiency of the LTX-2.3 series from Lightricks Want to try the fast open-source integration? Check it here on Diffusers: Want to try the fastest video generation endpoint? Check p-video here on API:

Pruna AI

25,748 Aufrufe • vor 2 Monaten

TurboEdit Instant text-based image editing discuss: We address the challenges of precise image inversion and disentangled image editing in the context of few-step diffusion models. We introduce an encoder based iterative inversion technique. The inversion network is conditioned on the input image and the reconstructed image from the previous step, allowing for correction of the next reconstruction towards the input image. We demonstrate that disentangled controls can be easily achieved in the few-step diffusion model by conditioning on an (automatically generated) detailed text prompt. To manipulate the inverted image, we freeze the noise maps and modify one attribute in the text prompt (either manually or via instruction based editing driven by an LLM), resulting in the generation of a new image similar to the input image with only one attribute changed. It can further control the editing strength and accept instructive text prompt. Our approach facilitates realistic text-guided image edits in real-time, requiring only 8 number of functional evaluations (NFEs) in inversion (one-time cost) and 4 NFEs per edit. Our method is not only fast, but also significantly outperforms state-of-the-art multi-step diffusion editing techniques.
0:22

TurboEdit Instant text-based image editing discuss: We address the challenges of precise image inversion and disentangled image editing in the context of few-step diffusion models. We introduce an encoder based iterative inversion technique. The inversion network is conditioned on the input image and the reconstructed image from the previous step, allowing for correction of the next reconstruction towards the input image. We demonstrate that disentangled controls can be easily achieved in the few-step diffusion model by conditioning on an (automatically generated) detailed text prompt. To manipulate the inverted image, we freeze the noise maps and modify one attribute in the text prompt (either manually or via instruction based editing driven by an LLM), resulting in the generation of a new image similar to the input image with only one attribute changed. It can further control the editing strength and accept instructive text prompt. Our approach facilitates realistic text-guided image edits in real-time, requiring only 8 number of functional evaluations (NFEs) in inversion (one-time cost) and 4 NFEs per edit. Our method is not only fast, but also significantly outperforms state-of-the-art multi-step diffusion editing techniques.

AK

16,062 Aufrufe • vor 2 Jahren

We've officially released and open-sourced HunyuanImage 2.1, our latest text-to-image model. The new model delivers on our commitment to balancing performance and quality. With native 2K image generation, HunyuanImage 2.1 is an advanced open-source text-to-image model.🎨 ✨ New in 2.1: 🔹Advanced Semantics: Supports ultra-long and complex prompts of up to 1000 tokens, and precisely controls the generation of multiple subjects in a single image. 🔹Precise Chinese and English Text Rendering with seamless image–text integration: The model naturally integrates text into images, making it suitable for a wide range of applications such as product covers, illustrations, and poster design to meet the needs of various fields. 🔹Rich Styles and High Aesthetic: Capable of generating images in various styles—including photorealistic portraits, comics, and vinyl figures—it delivers outstanding visual appeal and artistic quality. 🔹High-Quality Generation: Efficiently produces ultra-high-definition (2K) images in the same time other models take to generate a 1K image. HunyuanImage 2.1 uses two text encoders: a multimodal large language model (MLLM) to improve the model's image and text alignment capabilities, and a multi-language character-aware encoder to improve text rendering capabilities. The model is a single- and double-stream diffusion transformer with 17B parameters. We've also open-sourced the weights of the the accelerated version with meanflow which reduces inference steps from 100 to just 8, and PromptEnhancer, the first industrial-grade rewriting model that enhances your prompts for more nuanced and expressive image generation. Now, creators turn complex ideas—like posters with slogans or multi-panel comics—into visuals faster than ever. We’re just getting started. Stay tuned for our native multimodal image generation model coming soon. 🌐Website: 🔗Github: 🤗Hugging Face: ✨Hugging Face Demo:
1:39

We've officially released and open-sourced HunyuanImage 2.1, our latest text-to-image model. The new model delivers on our commitment to balancing performance and quality. With native 2K image generation, HunyuanImage 2.1 is an advanced open-source text-to-image model.🎨 ✨ New in 2.1: 🔹Advanced Semantics: Supports ultra-long and complex prompts of up to 1000 tokens, and precisely controls the generation of multiple subjects in a single image. 🔹Precise Chinese and English Text Rendering with seamless image–text integration: The model naturally integrates text into images, making it suitable for a wide range of applications such as product covers, illustrations, and poster design to meet the needs of various fields. 🔹Rich Styles and High Aesthetic: Capable of generating images in various styles—including photorealistic portraits, comics, and vinyl figures—it delivers outstanding visual appeal and artistic quality. 🔹High-Quality Generation: Efficiently produces ultra-high-definition (2K) images in the same time other models take to generate a 1K image. HunyuanImage 2.1 uses two text encoders: a multimodal large language model (MLLM) to improve the model's image and text alignment capabilities, and a multi-language character-aware encoder to improve text rendering capabilities. The model is a single- and double-stream diffusion transformer with 17B parameters. We've also open-sourced the weights of the the accelerated version with meanflow which reduces inference steps from 100 to just 8, and PromptEnhancer, the first industrial-grade rewriting model that enhances your prompts for more nuanced and expressive image generation. Now, creators turn complex ideas—like posters with slogans or multi-panel comics—into visuals faster than ever. We’re just getting started. Stay tuned for our native multimodal image generation model coming soon. 🌐Website: 🔗Github: 🤗Hugging Face: ✨Hugging Face Demo:

Tencent Hy

89,392 Aufrufe • vor 1 Jahr

InstantDrag Improving Interactivity in Drag-based Image Editing discuss: Drag-based image editing has recently gained popularity for its interactivity and precision. However, despite the ability of text-to-image models to generate samples within a second, drag editing still lags behind due to the challenge of accurately reflecting user interaction while maintaining image content. Some existing approaches rely on computationally intensive per-image optimization or intricate guidance-based methods, requiring additional inputs such as masks for movable regions and text prompts, thereby compromising the interactivity of the editing process. We introduce InstantDrag, an optimization-free pipeline that enhances interactivity and speed, requiring only an image and a drag instruction as input. InstantDrag consists of two carefully designed networks: a drag-conditioned optical flow generator (FlowGen) and an optical flow-conditioned diffusion model (FlowDiffusion). InstantDrag learns motion dynamics for drag-based image editing in real-world video datasets by decomposing the task into motion generation and motion-conditioned image generation. We demonstrate InstantDrag's capability to perform fast, photo-realistic edits without masks or text prompts through experiments on facial video datasets and general scenes. These results highlight the efficiency of our approach in handling drag-based image editing, making it a promising solution for interactive, real-time applications.
0:46

InstantDrag Improving Interactivity in Drag-based Image Editing discuss: Drag-based image editing has recently gained popularity for its interactivity and precision. However, despite the ability of text-to-image models to generate samples within a second, drag editing still lags behind due to the challenge of accurately reflecting user interaction while maintaining image content. Some existing approaches rely on computationally intensive per-image optimization or intricate guidance-based methods, requiring additional inputs such as masks for movable regions and text prompts, thereby compromising the interactivity of the editing process. We introduce InstantDrag, an optimization-free pipeline that enhances interactivity and speed, requiring only an image and a drag instruction as input. InstantDrag consists of two carefully designed networks: a drag-conditioned optical flow generator (FlowGen) and an optical flow-conditioned diffusion model (FlowDiffusion). InstantDrag learns motion dynamics for drag-based image editing in real-world video datasets by decomposing the task into motion generation and motion-conditioned image generation. We demonstrate InstantDrag's capability to perform fast, photo-realistic edits without masks or text prompts through experiments on facial video datasets and general scenes. These results highlight the efficiency of our approach in handling drag-based image editing, making it a promising solution for interactive, real-time applications.

AK

71,232 Aufrufe • vor 2 Jahren

P-Image-Upscale is the fastest and cheapest image upscaler in the world: supporting outputs up to 128 MP under 1 seconds. A couple of weeks ago we released p-image-upscale, and it now works better than ever. It’s the fastest image upscaler in the world, supporting outputs up to 128 MP while keeping pricing simple and predictable: - $0.005/image for 1–4 MP - $0.01/image for 4–8 MP - $0.02/image for 8–16 MP - $0.04/image for 16–32 MP - $0.06/image for 32–64 MP - $0.12/image for 64–128 MP That means you can go from low-res input to production-ready output with extreme speed, while preserving detail and keeping costs easy to understand. Available on: - Pruna AI | - each::labs | - inference shell | - Replicate | - Runware | - Segmind | - WaveSpeedAI | - Wiro AI — Ship AI Faster | If you’re building workflows where image quality, price, latency, and scale all matter, p-image-upscale is built for you.
0:20

P-Image-Upscale is the fastest and cheapest image upscaler in the world: supporting outputs up to 128 MP under 1 seconds. A couple of weeks ago we released p-image-upscale, and it now works better than ever. It’s the fastest image upscaler in the world, supporting outputs up to 128 MP while keeping pricing simple and predictable: - $0.005/image for 1–4 MP - $0.01/image for 4–8 MP - $0.02/image for 8–16 MP - $0.04/image for 16–32 MP - $0.06/image for 32–64 MP - $0.12/image for 64–128 MP That means you can go from low-res input to production-ready output with extreme speed, while preserving detail and keeping costs easy to understand. Available on: - Pruna AI | - each::labs | - inference shell | - Replicate | - Runware | - Segmind | - WaveSpeedAI | - Wiro AI — Ship AI Faster | If you’re building workflows where image quality, price, latency, and scale all matter, p-image-upscale is built for you.

Pruna AI

24,089 Aufrufe • vor 4 Monaten

We’re excited to announce the release and open-source of HunyuanImage 3.0 — the largest and most powerful open-source text-to-image model to date, with over 80 billion total parameters, of which 13 billion are activated per token during inference.The effect is completely comparable to the industry’s flagship closed-source model.🚀🚀🚀 HunyuanImage 3.0 originates from our internally developed native multimodal large language model, with fine-tuning and post-training focused on text-to-image generation. This unique foundation gives the model a powerful set of capabilities: ✅Reason with world knowledge ✅Understand complex, thousand-word prompts ✅Generate precise text within images Different from traditional DiT architecture image generation models, HunyuanImage 3.0’s MoE architecture uses a Transfusion-based approach to deeply couple Diffusion and LLM training for a single, powerful system. Built on Hunyuan-A13B, HunyuanImage 3.0 was trained on a massive dataset: 5 billion image-text pairs, video frames, interleaved image-text data, and 6 trillion tokens of text corpora. This hybrid training across multimodal generation, understanding, and LLM capabilities allows the model to seamlessly integrate multiple tasks. Whether you're an illustrator, designer, or creator, this is built to slash your workflow from hours to minutes. HunyuanImage 3.0 can generate intricate text, detailed comics, expressive emojis, and lively, engaging illustrations for educational content. The current release focuses solely on text-to-image generation and future updates will include image-to-image, image editing, multi-turn interaction, and more. 👉🏻Try it now: 🔗GitHub: 🤗Hugging Face:
2:09

We’re excited to announce the release and open-source of HunyuanImage 3.0 — the largest and most powerful open-source text-to-image model to date, with over 80 billion total parameters, of which 13 billion are activated per token during inference.The effect is completely comparable to the industry’s flagship closed-source model.🚀🚀🚀 HunyuanImage 3.0 originates from our internally developed native multimodal large language model, with fine-tuning and post-training focused on text-to-image generation. This unique foundation gives the model a powerful set of capabilities: ✅Reason with world knowledge ✅Understand complex, thousand-word prompts ✅Generate precise text within images Different from traditional DiT architecture image generation models, HunyuanImage 3.0’s MoE architecture uses a Transfusion-based approach to deeply couple Diffusion and LLM training for a single, powerful system. Built on Hunyuan-A13B, HunyuanImage 3.0 was trained on a massive dataset: 5 billion image-text pairs, video frames, interleaved image-text data, and 6 trillion tokens of text corpora. This hybrid training across multimodal generation, understanding, and LLM capabilities allows the model to seamlessly integrate multiple tasks. Whether you're an illustrator, designer, or creator, this is built to slash your workflow from hours to minutes. HunyuanImage 3.0 can generate intricate text, detailed comics, expressive emojis, and lively, engaging illustrations for educational content. The current release focuses solely on text-to-image generation and future updates will include image-to-image, image editing, multi-turn interaction, and more. 👉🏻Try it now: 🔗GitHub: 🤗Hugging Face:

Tencent Hy

413,096 Aufrufe • vor 1 Jahr

As I promised yesterday, I'll briefly explain LoRA training and share a workflow I made so you can do it quickly. First, let me answer a very common question: 'Why train LoRAs when we have such advanced models?' Even though we have incredibly advanced models now (like NBP), we still can't always get them to do specific things we want. Simplest example: the spritesheet LoRA I made the other day. I generated 1000 images with Nano Banana and only 100 were what I wanted. The LoRA I trained using those 100 images gives me nearly 100% consistent results. Second point is cost and speed. With LoRA, we can cut costs by 4-5x. And while doing that, we're generating 4-5x faster. How many images do you need for a good LoRA? This depends on your LoRA's complexity. For example, when I training the spritesheet LoRA, even though I used 100 images, I didn't include buildings in the training data, so this LoRA doesn't work for buildings. So think about your LoRA's use cases and add examples for as many use cases as possible to improve quality. What are paired images and how to train LoRAs for image-editing? When training LoRAs for image editing on fal, we call each edit example paired images - one with _start suffix, one with _end suffix. For example, if you're training a background remove LoRA, the unedited original photo will be your '_start' image. The image with background removed will be the '_end' image. Simply put: images we want to edit or use as reference get _start, target images we want to achieve get '_end'. Important: save both images with the same name. Like image332_start.jpg and image332_end.jpg. This way the system knows which images pair together. What about training LoRAs for models with multiple image inputs? Same logic. We still use _start and _end suffixes, but with one difference. Since there are multiple input images, we can number them: _start, _start1, _start2. Example: start images, 1st image = Woman portrait (image35_start.jpg) 2nd image = Glasses photo (image35_start1.jpg) 3rd image = Hat photo (image35_start2.jpg) Output image = portrait of woman wearing glasses and hat (image35_end.jpg) Can we do more detailed captioning? Yes. Similarly, you can improve training quality by creating a txt file for each set with the caption inside. Example: create image35.txt and write: 'Recreate the image by putting the glasses from the second image and the hat from the third image on the woman in the first image.' What are Steps? How many should I use? What's Learning Rate? Steps determines how many times the model sees and processes your training data (your images). Each step, the model learns a bit more. But as steps increase, so does the risk of overfitting. So there's no real default. But for a simpler LoRA with 20 paired images, 1000 steps is ideal. Here's a metaphor for the Steps and Learning Rate relationship: Imagine you have a balloon. Our goal is to inflate it to the optimal size. Steps = How many times we blow into the balloon Learning rate = How hard we blow each time If we blow too softly, we need to blow many more times. If we blow too hard, we risk popping it quickly and can't reach optimal size. Of course training won't explode, but it won't work as intended because it wasn't trained optimally. Training's done, now what? Once training's complete, you'll have a safetensors file. Every model you train on fal has a LoRA inference endpoint. In that inference, add your safetensors file link to the LoRA url input, and you can use your LoRA. Thanks for the read! The workflow in the video: If I forgot anything, let me know in the replies.
1:06

As I promised yesterday, I'll briefly explain LoRA training and share a workflow I made so you can do it quickly. First, let me answer a very common question: 'Why train LoRAs when we have such advanced models?' Even though we have incredibly advanced models now (like NBP), we still can't always get them to do specific things we want. Simplest example: the spritesheet LoRA I made the other day. I generated 1000 images with Nano Banana and only 100 were what I wanted. The LoRA I trained using those 100 images gives me nearly 100% consistent results. Second point is cost and speed. With LoRA, we can cut costs by 4-5x. And while doing that, we're generating 4-5x faster. How many images do you need for a good LoRA? This depends on your LoRA's complexity. For example, when I training the spritesheet LoRA, even though I used 100 images, I didn't include buildings in the training data, so this LoRA doesn't work for buildings. So think about your LoRA's use cases and add examples for as many use cases as possible to improve quality. What are paired images and how to train LoRAs for image-editing? When training LoRAs for image editing on fal, we call each edit example paired images - one with _start suffix, one with _end suffix. For example, if you're training a background remove LoRA, the unedited original photo will be your '_start' image. The image with background removed will be the '_end' image. Simply put: images we want to edit or use as reference get _start, target images we want to achieve get '_end'. Important: save both images with the same name. Like image332_start.jpg and image332_end.jpg. This way the system knows which images pair together. What about training LoRAs for models with multiple image inputs? Same logic. We still use _start and _end suffixes, but with one difference. Since there are multiple input images, we can number them: _start, _start1, _start2. Example: start images, 1st image = Woman portrait (image35_start.jpg) 2nd image = Glasses photo (image35_start1.jpg) 3rd image = Hat photo (image35_start2.jpg) Output image = portrait of woman wearing glasses and hat (image35_end.jpg) Can we do more detailed captioning? Yes. Similarly, you can improve training quality by creating a txt file for each set with the caption inside. Example: create image35.txt and write: 'Recreate the image by putting the glasses from the second image and the hat from the third image on the woman in the first image.' What are Steps? How many should I use? What's Learning Rate? Steps determines how many times the model sees and processes your training data (your images). Each step, the model learns a bit more. But as steps increase, so does the risk of overfitting. So there's no real default. But for a simpler LoRA with 20 paired images, 1000 steps is ideal. Here's a metaphor for the Steps and Learning Rate relationship: Imagine you have a balloon. Our goal is to inflate it to the optimal size. Steps = How many times we blow into the balloon Learning rate = How hard we blow each time If we blow too softly, we need to blow many more times. If we blow too hard, we risk popping it quickly and can't reach optimal size. Of course training won't explode, but it won't work as intended because it wasn't trained optimally. Training's done, now what? Once training's complete, you'll have a safetensors file. Every model you train on fal has a LoRA inference endpoint. In that inference, add your safetensors file link to the LoRA url input, and you can use your LoRA. Thanks for the read! The workflow in the video: If I forgot anything, let me know in the replies.

ilker

15,192 Aufrufe • vor 8 Monaten

Google Pics — our image creation and editing tool in Google Workspace — is rolling out to Google AI Pro and Ultra subscribers and most Workspace business customers. Built on our Nano Banana image generation and editing model, Pics gives you the control you need to create whatever you're imagining — like designing a poster, producing social media posts, or editing digital illustrations. It also includes a suite of fine-tuned editing capabilities that let you generate images and make precise edits rather than starting from scratch.
1:38

Google Pics — our image creation and editing tool in Google Workspace — is rolling out to Google AI Pro and Ultra subscribers and most Workspace business customers. Built on our Nano Banana image generation and editing model, Pics gives you the control you need to create whatever you're imagining — like designing a poster, producing social media posts, or editing digital illustrations. It also includes a suite of fine-tuned editing capabilities that let you generate images and make precise edits rather than starting from scratch.

Google

248,622 Aufrufe • vor 17 Tagen

Today, we introduce HunyuanImage 3.0-Instruct, a native multimodal model focusing on image-editing by integrating visual understanding with precise image synthesis! 🚀 It understands input images and reasons before generating images. Built on an 80B-parameter MoE architecture (13B activated), it natively unifies deep multimodal comprehension and high-fidelity generation. 🧠 A "Thinking" Model with Native CoT & MixGRPO: The model doesn’t just execute commands, it processes them through a Native Chain-of-Thought (CoT) schema. Enhanced by our self-developed MixGRPO algorithm, it reasons through complex instructions to achieve flawless intent alignment and human-preference consistency. 🎨 Precise Editing & Multi-Image Fusion: The model enables accurate image editing by adding, removing, or modifying elements while keeping non-target areas perfectly intact. It also excels at seamless multi-image fusion, synthesizing complex scenes by extracting and blending elements from multiple sources into a unified, consistent output. 🏆 SOTA Performance: HunyuanImage 3.0-Instruct sets a new benchmark in visual quality and alignment, delivering performance that matches leading proprietary models. We aim to enable the community to explore new ideas with a state-of-the-art foundation model, fostering a dynamic and vibrant image generation ecosystem. 🛠️🎨 💻Try it at (PC only):
1:12

Today, we introduce HunyuanImage 3.0-Instruct, a native multimodal model focusing on image-editing by integrating visual understanding with precise image synthesis! 🚀 It understands input images and reasons before generating images. Built on an 80B-parameter MoE architecture (13B activated), it natively unifies deep multimodal comprehension and high-fidelity generation. 🧠 A "Thinking" Model with Native CoT & MixGRPO: The model doesn’t just execute commands, it processes them through a Native Chain-of-Thought (CoT) schema. Enhanced by our self-developed MixGRPO algorithm, it reasons through complex instructions to achieve flawless intent alignment and human-preference consistency. 🎨 Precise Editing & Multi-Image Fusion: The model enables accurate image editing by adding, removing, or modifying elements while keeping non-target areas perfectly intact. It also excels at seamless multi-image fusion, synthesizing complex scenes by extracting and blending elements from multiple sources into a unified, consistent output. 🏆 SOTA Performance: HunyuanImage 3.0-Instruct sets a new benchmark in visual quality and alignment, delivering performance that matches leading proprietary models. We aim to enable the community to explore new ideas with a state-of-the-art foundation model, fostering a dynamic and vibrant image generation ecosystem. 🛠️🎨 💻Try it at (PC only):

Tencent Hy

126,668 Aufrufe • vor 8 Monaten

Grok Imagine API now supports built-in video editing on top of powerful video generation You can feed in your own videos and edit them directly with Grok Imagine, transforming or enhancing footage without traditional timeline editing Developers can generate high-quality videos at scale with fast iteration and simple integration into existing workflows One API does it all: • Text-to-video • Image-to-video • Edit existing videos directly • Replace, remove, or add objects inside a video • Built-in audio sync for every clip Turn your ideas into finished videos – at scale, through a single API
0:43

Grok Imagine API now supports built-in video editing on top of powerful video generation You can feed in your own videos and edit them directly with Grok Imagine, transforming or enhancing footage without traditional timeline editing Developers can generate high-quality videos at scale with fast iteration and simple integration into existing workflows One API does it all: • Text-to-video • Image-to-video • Edit existing videos directly • Replace, remove, or add objects inside a video • Built-in audio sync for every clip Turn your ideas into finished videos – at scale, through a single API

X Freeze

19,054 Aufrufe • vor 7 Monaten

most AI APIs either cost too much or lock the best models behind expensive plans. then i came across Agnes AI. it’s one of the few platforms giving developers access to text, image, and video AI models through a single API. their ecosystem gives you direct access to top-tier models like - Agnes-2.0-Flash for text - Agnes-Image-2.0 for image - Agnes-Video-V2.0 for video Instead of managing different providers for different tasks, everything works from one place. what stood out to me: > one API for text, image, and video generation > OpenAI-compatible API, so integration is straightforward > built for speed and scalability > free access for developers to start building > useful for everything from chatbots to image generation and content automation if you’re building AI products, it’s worth taking a look while it’s still flying under the radar. sometimes the biggest advantage is finding great tools before everyone else does. check out Agnes AI here, get free access for developers : and if you’re building with it, join the developer community here: Agnes AI #AgnesAI #AgnesAPI #BuildWithAgnes #MultimodalAI #AIforDevs #DeveloperAPI #AIParity
1:37

most AI APIs either cost too much or lock the best models behind expensive plans. then i came across Agnes AI. it’s one of the few platforms giving developers access to text, image, and video AI models through a single API. their ecosystem gives you direct access to top-tier models like - Agnes-2.0-Flash for text - Agnes-Image-2.0 for image - Agnes-Video-V2.0 for video Instead of managing different providers for different tasks, everything works from one place. what stood out to me: > one API for text, image, and video generation > OpenAI-compatible API, so integration is straightforward > built for speed and scalability > free access for developers to start building > useful for everything from chatbots to image generation and content automation if you’re building AI products, it’s worth taking a look while it’s still flying under the radar. sometimes the biggest advantage is finding great tools before everyone else does. check out Agnes AI here, get free access for developers : and if you’re building with it, join the developer community here: Agnes AI #AgnesAI #AgnesAPI #BuildWithAgnes #MultimodalAI #AIforDevs #DeveloperAPI #AIParity

MIKE

14,591 Aufrufe • vor 2 Monaten

ComfyUI Tutorial for Beginners: Text-to-Image & Image-to-Image Workflows One graph, two workflows and Module 2 is where you build it node by node instead of loading someone else's. We start from a blank canvas and wire every connection, then change a single value and turn the same graph into image-to-image. What you'll learn: → The 5 core nodes, connected one at a time — checkpoint, CLIP encode, KSampler, empty latent, VAE decode → Why the data flows the way it does, not just where the wires go → Every KSampler setting: seed, steps, CFG, sampler, scheduler → Denoise from 0.9 to 0.0 — the one value that decides how much of your source survives → Locking seeds and generating takes instead of chasing one perfect image → Preview vs. save, and why every image carries the graph that made it → Reopen any result in your asset library as a live workflow → The 6 errors that break Comfy most often → How to read a red node like a sentence Click the links below to run the workflow ⬇️
37:05

ComfyUI Tutorial for Beginners: Text-to-Image & Image-to-Image Workflows One graph, two workflows and Module 2 is where you build it node by node instead of loading someone else's. We start from a blank canvas and wire every connection, then change a single value and turn the same graph into image-to-image. What you'll learn: → The 5 core nodes, connected one at a time — checkpoint, CLIP encode, KSampler, empty latent, VAE decode → Why the data flows the way it does, not just where the wires go → Every KSampler setting: seed, steps, CFG, sampler, scheduler → Denoise from 0.9 to 0.0 — the one value that decides how much of your source survives → Locking seeds and generating takes instead of chasing one perfect image → Preview vs. save, and why every image carries the graph that made it → Reopen any result in your asset library as a live workflow → The 6 errors that break Comfy most often → How to read a red node like a sentence Click the links below to run the workflow ⬇️

ComfyUI

58,853 Aufrufe • vor 26 Tagen

Meet Qwen-VLo, your AI creative engine: • Concept-to-Polish: Turn rough sketches or text prompts into high-res visuals • On-the-Fly Edits: Refine product shots, adjust layouts or styles with simple commands • Global-Ready: Generate image in multiple languages • Progressive Generation: Build complex scenes step-by-step Perfect for designers, marketers, educators—and anyone who wants to bring ideas to life. 👉 Try it: 📖 Details:
1:27

Meet Qwen-VLo, your AI creative engine: • Concept-to-Polish: Turn rough sketches or text prompts into high-res visuals • On-the-Fly Edits: Refine product shots, adjust layouts or styles with simple commands • Global-Ready: Generate image in multiple languages • Progressive Generation: Build complex scenes step-by-step Perfect for designers, marketers, educators—and anyone who wants to bring ideas to life. 👉 Try it: 📖 Details:

Qwen

149,929 Aufrufe • vor 1 Jahr

Announcing Gemini 2.5 Flash Image! (Nano Banana) from Google DeepMind. This model brings you state of the art image generation! Here's 7m of demos. Today marks a leap forward in AI-powered image creativity with the launch of stronger native image generation and editing in Gemini 2.5 Flash. Here's what's new: 👤 Consistent character design: Maintain a character's identity across multiple images, changing their outfits, poses, and scenes without losing their core look. 🎨 Creative composition: Seamlessly merge elements from up to three different images into a single, cohesive masterpiece. The potential for surrealist art and unique compositions is endless. ✍️ Targeted edits with natural language: Go beyond the prompt. You can now make highly specific edits like "change the background," "restore this faded photo," or "edit the character's outfit" with simple commands. ✨ Design & identity adaptation: Apply a specific style, texture, or pattern from one image to another, unlocking new possibilities for fashion, product design, and architectural visualization. I've put together a ~7-minute video demo to walk you through these features and show just how powerful they are in action. We can't wait to see what you all create! Try the model out in Google AI Studio, Gemini app or the Gemini API.
7:27

Announcing Gemini 2.5 Flash Image! (Nano Banana) from Google DeepMind. This model brings you state of the art image generation! Here's 7m of demos. Today marks a leap forward in AI-powered image creativity with the launch of stronger native image generation and editing in Gemini 2.5 Flash. Here's what's new: 👤 Consistent character design: Maintain a character's identity across multiple images, changing their outfits, poses, and scenes without losing their core look. 🎨 Creative composition: Seamlessly merge elements from up to three different images into a single, cohesive masterpiece. The potential for surrealist art and unique compositions is endless. ✍️ Targeted edits with natural language: Go beyond the prompt. You can now make highly specific edits like "change the background," "restore this faded photo," or "edit the character's outfit" with simple commands. ✨ Design & identity adaptation: Apply a specific style, texture, or pattern from one image to another, unlocking new possibilities for fashion, product design, and architectural visualization. I've put together a ~7-minute video demo to walk you through these features and show just how powerful they are in action. We can't wait to see what you all create! Try the model out in Google AI Studio, Gemini app or the Gemini API.

Addy Osmani

159,178 Aufrufe • vor 1 Jahr

📢📢 𝐏𝐞𝐫𝐜𝐇𝐞𝐚𝐝: 𝐏𝐞𝐫𝐜𝐞𝐩𝐭𝐮𝐚𝐥 𝐇𝐞𝐚𝐝 𝐌𝐨𝐝𝐞𝐥 𝐟𝐨𝐫 𝐒𝐢𝐧𝐠𝐥𝐞-𝐈𝐦𝐚𝐠𝐞 𝟑𝐃 𝐇𝐞𝐚𝐝 𝐑𝐞𝐜𝐨𝐧𝐬𝐭𝐫𝐮𝐜𝐭𝐢𝐨𝐧 & 𝐄𝐝𝐢𝐭𝐢𝐧𝐠📢📢 PercHead reconstructs realistic 3D heads from a single image and enables disentangled 3D editing via geometric controls and style inputs from images or text. At its core is a generalized 3D head decoder trained with perceptual supervision from DINOv2 and SAM 2.1. We find that our new perceptual loss formulation improves reconstruction fidelity compared to commonly-used methods such as LPIPS. Our trained reconstruction model is able to generate 3D-consistent heads from a single input image. Even with challenging side-view inputs, the model robustly infers missing regions for a coherent, high-fidelity output. In addition, our architecture seamlessly adapts to downstream tasks: by swapping the encoder, we can transform the model into a disentangled 3D editing pipeline. In this scenario, we can control geometry through - potentially hand-drawn - segmentation maps, and condition style via image or text prompt. We also provide an interactive GUI to enable the exploration of our editing pipeline. 🌍 📽️ Great work by Antonio Oroz and Tobias Kirschstein
0:30

📢📢 𝐏𝐞𝐫𝐜𝐇𝐞𝐚𝐝: 𝐏𝐞𝐫𝐜𝐞𝐩𝐭𝐮𝐚𝐥 𝐇𝐞𝐚𝐝 𝐌𝐨𝐝𝐞𝐥 𝐟𝐨𝐫 𝐒𝐢𝐧𝐠𝐥𝐞-𝐈𝐦𝐚𝐠𝐞 𝟑𝐃 𝐇𝐞𝐚𝐝 𝐑𝐞𝐜𝐨𝐧𝐬𝐭𝐫𝐮𝐜𝐭𝐢𝐨𝐧 & 𝐄𝐝𝐢𝐭𝐢𝐧𝐠📢📢 PercHead reconstructs realistic 3D heads from a single image and enables disentangled 3D editing via geometric controls and style inputs from images or text. At its core is a generalized 3D head decoder trained with perceptual supervision from DINOv2 and SAM 2.1. We find that our new perceptual loss formulation improves reconstruction fidelity compared to commonly-used methods such as LPIPS. Our trained reconstruction model is able to generate 3D-consistent heads from a single input image. Even with challenging side-view inputs, the model robustly infers missing regions for a coherent, high-fidelity output. In addition, our architecture seamlessly adapts to downstream tasks: by swapping the encoder, we can transform the model into a disentangled 3D editing pipeline. In this scenario, we can control geometry through - potentially hand-drawn - segmentation maps, and condition style via image or text prompt. We also provide an interactive GUI to enable the exploration of our editing pipeline. 🌍 📽️ Great work by Antonio Oroz and Tobias Kirschstein

Matthias Niessner

18,903 Aufrufe • vor 10 Monaten

[CLIP] by Hand ✍️ The CLIP (Contrastive Language–Image Pre-training) model, a groundbreaking work by OpenAI, redefines the intersection of computer vision and natural language processing. It is the basis of all the multi-modal foundation models we see today. How does CLIP work? Goal: 🟨 Learn a shared embedding space for text and image [1] Given ↳ A mini batch of 3 text-image pairs ↳ OpenAI used 400 million text-image pairs to train its original CLIP model. Process 1st pair: "big table" [2] 🟪 Text → 2 Vectors (3D) ↳ Look up word embedding vectors using word2vec. [3] 🟩 Image → 2 Vectors (4D) ↳ Divide the image into two patches. ↳ Flatten each patch [4] Process other pairs ↳ Repeat [2]-[3] [5] 🟪 Text Encoder & 🟩 Image Encoder ↳ Encode input vectors into feature vectors ↳ Here, both encoders are simple one layer perceptron (linear + ReLU) ↳ In practice, the encoders are usually transformer models. [6] 🟪 🟩 Mean Pooling: 2 → 1 vector ↳ Average 2 feature vectors into a single vector by averaging across the columns ↳ The goal is to have one vector to represent each image or text [7] 🟪 🟩 -> 🟨 Projection ↳ Note that the text and image feature vectors from the encoders have different dimensions (3D vs. 4D). ↳ Use a linear layer to project image and text vectors to a 2D shared embedding space. 🏋️ Contrastive Pre-training 🏋️ [8] Prepare for MatMul ↳ Copy text vectors (T1,T2,T3) ↳ Copy the transpose of image vectors (I1,I2,I3) ↳ They are all in the 2D shared embedding space. [9] 🟦 MatMul ↳ Multiply T and I matrices. ↳ This is equivalent to taking dot product between every pair of image and text vectors. ↳ The purpose is to use dot product to estimate the similarity between a pair of image-text. [10] 🟦 Softmax: e^x ↳ Raise e to the power of the number in each cell ↳ To simplify hand calculation, we approximate e^□ with 3^□. [11] 🟦 Softmax: ∑ ↳ Sum each row for 🟩 image→🟪 text ↳ Sum each column for 🟪 text→ 🟩 image [12] 🟦 Softmax: 1 / sum ↳ Divide each element by the column sum to obtain a similarity matrix for 🟪 text→🟩 image ↳ Divide each element by the row sum to obtain a similarity matrix for 🟩 image→🟪 text [13] 🟥 Loss Gradients ↳ The "Targets" for the similarity matrices are Identity Matrices. ↳ Why? If I and T come from the same pair (i=j), we want the highest value, which is 1, and 0 otherwise. ↳ Apply the simple equation of [Similarity - Target] to compute gradients of for both directions. ↳ Why so simple? Because when Softmax and Cross-Entropy Loss are used together, the math magically works out that way. ↳ These gradients kick off the backpropagation process to update weights and biases of the encoders and projection layers (red borders).
1:20

[CLIP] by Hand ✍️ The CLIP (Contrastive Language–Image Pre-training) model, a groundbreaking work by OpenAI, redefines the intersection of computer vision and natural language processing. It is the basis of all the multi-modal foundation models we see today. How does CLIP work? Goal: 🟨 Learn a shared embedding space for text and image [1] Given ↳ A mini batch of 3 text-image pairs ↳ OpenAI used 400 million text-image pairs to train its original CLIP model. Process 1st pair: "big table" [2] 🟪 Text → 2 Vectors (3D) ↳ Look up word embedding vectors using word2vec. [3] 🟩 Image → 2 Vectors (4D) ↳ Divide the image into two patches. ↳ Flatten each patch [4] Process other pairs ↳ Repeat [2]-[3] [5] 🟪 Text Encoder & 🟩 Image Encoder ↳ Encode input vectors into feature vectors ↳ Here, both encoders are simple one layer perceptron (linear + ReLU) ↳ In practice, the encoders are usually transformer models. [6] 🟪 🟩 Mean Pooling: 2 → 1 vector ↳ Average 2 feature vectors into a single vector by averaging across the columns ↳ The goal is to have one vector to represent each image or text [7] 🟪 🟩 -> 🟨 Projection ↳ Note that the text and image feature vectors from the encoders have different dimensions (3D vs. 4D). ↳ Use a linear layer to project image and text vectors to a 2D shared embedding space. 🏋️ Contrastive Pre-training 🏋️ [8] Prepare for MatMul ↳ Copy text vectors (T1,T2,T3) ↳ Copy the transpose of image vectors (I1,I2,I3) ↳ They are all in the 2D shared embedding space. [9] 🟦 MatMul ↳ Multiply T and I matrices. ↳ This is equivalent to taking dot product between every pair of image and text vectors. ↳ The purpose is to use dot product to estimate the similarity between a pair of image-text. [10] 🟦 Softmax: e^x ↳ Raise e to the power of the number in each cell ↳ To simplify hand calculation, we approximate e^□ with 3^□. [11] 🟦 Softmax: ∑ ↳ Sum each row for 🟩 image→🟪 text ↳ Sum each column for 🟪 text→ 🟩 image [12] 🟦 Softmax: 1 / sum ↳ Divide each element by the column sum to obtain a similarity matrix for 🟪 text→🟩 image ↳ Divide each element by the row sum to obtain a similarity matrix for 🟩 image→🟪 text [13] 🟥 Loss Gradients ↳ The "Targets" for the similarity matrices are Identity Matrices. ↳ Why? If I and T come from the same pair (i=j), we want the highest value, which is 1, and 0 otherwise. ↳ Apply the simple equation of [Similarity - Target] to compute gradients of for both directions. ↳ Why so simple? Because when Softmax and Cross-Entropy Loss are used together, the math magically works out that way. ↳ These gradients kick off the backpropagation process to update weights and biases of the encoders and projection layers (red borders).

Tom Yeh

67,896 Aufrufe • vor 2 Jahren

StyleDrop: Text-to-Image Generation in Any Style introduce StyleDrop, a method that enables the synthesis of images that faithfully follow a specific style using a text-to-image model. The proposed method is extremely versatile and captures nuances and details of a user-provided style, such as color schemes, shading, design patterns, and local and global effects. It efficiently learns a new style by fine-tuning very few trainable parameters (less than 1% of total model parameters) and improving the quality via iterative training with either human or automated feedback. Better yet, StyleDrop is able to deliver impressive results even when the user supplies only a single image that specifies the desired style. An extensive study shows that, for the task of style tuning text-to-image models, StyleDrop implemented on Muse convincingly outperforms other methods, including DreamBooth and textual inversion on Imagen or Stable Diffusion. paper page:
1:56

StyleDrop: Text-to-Image Generation in Any Style introduce StyleDrop, a method that enables the synthesis of images that faithfully follow a specific style using a text-to-image model. The proposed method is extremely versatile and captures nuances and details of a user-provided style, such as color schemes, shading, design patterns, and local and global effects. It efficiently learns a new style by fine-tuning very few trainable parameters (less than 1% of total model parameters) and improving the quality via iterative training with either human or automated feedback. Better yet, StyleDrop is able to deliver impressive results even when the user supplies only a single image that specifies the desired style. An extensive study shows that, for the task of style tuning text-to-image models, StyleDrop implemented on Muse convincingly outperforms other methods, including DreamBooth and textual inversion on Imagen or Stable Diffusion. paper page:

AK

56,402 Aufrufe • vor 3 Jahren