Loading video...

Video Failed to Load

There was a problem loading this video. This could be due to a temporary network issue or the video might be unavailable.

㊗️Congrats on Lvmin Zhang’s (github@lllyasviel) latest project FramePack and thank you for using and recommending HunyuanVideo. 😀So happy to see innovations based on Hunyuan and we would like to see more. ▶️FramePack's Brief Intro and Showcases Attached: FramePack is a next-frame prediction neural network structure that generates videos progressively,... show more

Tencent HY

33,591 subscribers

210,162 views • 1 year ago •via X (Twitter)

Gaming Science & Technology Education

Anya Rossi• Live Now

Private livecam show

11 Comments

Rud_ai1 year ago

So is the generation now autoregressive?

OPEN1 year ago

From the wilderness to futuristic cities, your OP3N adventure starts here. Join our Discord channel and become part of the community!

Abu siddik1 year ago

Hey, Let's connect for collaboration

Hasan1 year ago

Fantastic work, Lvmin! Excited to see what you create next with HunyuanVideo.

Risichad 🦾1 year ago

Amazing new stuff

JuwalBose1 year ago

Amazing work.

Amira Zairi1 year ago

Great work!

Manish Kumar Shah1 year ago

Let's connect for collaboration

Sentient Matrix1 year ago

This looks very interesting, thanks for making it work on consumer hardware! Seems like a natural candidate for Pinokio @pean

Castani1 year ago

😀⚡️

Bandinopla1 year ago

this is insane, the frames are very loyal to the original one, wow. Good job!

Related Videos

Meta announces Movie Gen A Cast of Media Foundation Models We present Movie Gen, a cast of foundation models that generates high-quality, 1080p HD videos with different aspect ratios and synchronized audio. We also show additional capabilities such as precise instruction-based video editing and generation of personalized videos based on a user’s image. Our models set a new state-of-the-art on multiple tasks: text-to-video synthesis, video personalization, video editing, video-to-audio generation, and text-to-audio generation. Our largest video generation model is a 30B parameter transformer trained with a maximum context length of 73K video tokens, corresponding to a generated video of 16 seconds at 16 frames-per-second. We show multiple technical innovations and simplifications on the architecture, latent spaces, training objectives and recipes, data curation, evaluation protocols, parallelization techniques, and inference optimizations that allow us to reap the benefits of scaling pre-training data, model size, and training compute for training large scale media generation models. We hope this paper helps the research community to accelerate progress and innovation in media generation models

Meta announces Movie Gen A Cast of Media Foundation Models We present Movie Gen, a cast of foundation models that generates high-quality, 1080p HD videos with different aspect ratios and synchronized audio. We also show additional capabilities such as precise instruction-based video editing and generation of personalized videos based on a user’s image. Our models set a new state-of-the-art on multiple tasks: text-to-video synthesis, video personalization, video editing, video-to-audio generation, and text-to-audio generation. Our largest video generation model is a 30B parameter transformer trained with a maximum context length of 73K video tokens, corresponding to a generated video of 16 seconds at 16 frames-per-second. We show multiple technical innovations and simplifications on the architecture, latent spaces, training objectives and recipes, data curation, evaluation protocols, parallelization techniques, and inference optimizations that allow us to reap the benefits of scaling pre-training data, model size, and training compute for training large scale media generation models. We hope this paper helps the research community to accelerate progress and innovation in media generation models

AK

62,719 views • 1 year ago

Introducing The Matrix --- a foundation world model for generating infinite-length, hyper-realistic videos with real-time, frame-level control: - Infinite-length video generation - 720p high-quality rendering - Real-time, frame-level control at 16 FPS - Generalization to real-world video control 🔗Blog: 📄Paper: 💻Code & Playable Demo: Coming soon! Key Innovation: A brand new technique called the shift-window denoise process model, enabling auto-regressive generation for diffusion and consistency models in real-time. Special thanks to project leader Ruili Feng and the entire Matrix team for their dedication and hard work over the year-long project.

Hongyang Zhang

178,372 views • 1 year ago

🔗: i just wanted to say that youve been really important to EN. your legacy is so intertwined to the dna of our culture, so many EN jokes are just 🌹's jokes basically 🔗: youre like one of the funniest people in, maybe in the world and sometimes unintentionally. your ability to turn any situation into humour speaks to your positivity and sincerity 🔗: thats been like a shining beacon on us for so long, thank you for that. thank you for being a part of EN with me. its been an honor 🌹: see you next time! 🔗: see you later, i hope? byee 🌹: byeee 😭🥹

🔗: i just wanted to say that youve been really important to EN. your legacy is so intertwined to the dna of our culture, so many EN jokes are just 🌹's jokes basically 🔗: youre like one of the funniest people in, maybe in the world and sometimes unintentionally. your ability to turn any situation into humour speaks to your positivity and sincerity 🔗: thats been like a shining beacon on us for so long, thank you for that. thank you for being a part of EN with me. its been an honor 🌹: see you next time! 🔗: see you later, i hope? byee 🌹: byeee 😭🥹

luna 🧸

51,852 views • 11 months ago

🕹️We are excited to introduce "ChronoEdit: Towards Temporal Reasoning for Image Editing and World Simulation" ChronoEdit reframes image editing as a video generation task to encourage temporal consistency. It leverages a temporal reasoning stage that denoises with “video reasoning tokens” to "reason" on physically plausible edits. See the attached video for results. Project Page: Arxiv: Code and model are coming.

🕹️We are excited to introduce "ChronoEdit: Towards Temporal Reasoning for Image Editing and World Simulation" ChronoEdit reframes image editing as a video generation task to encourage temporal consistency. It leverages a temporal reasoning stage that denoises with “video reasoning tokens” to "reason" on physically plausible edits. See the attached video for results. Project Page: Arxiv: Code and model are coming.

Huan Ling

36,895 views • 9 months ago

Meta Tribe2 is a model that predicts your brain's response on videos so now you can edit videos with more clarity of how to compose them, what works and what doesn't and we built a free tool for you based on this model yes, totally free see details below

Meta Tribe2 is a model that predicts your brain's response on videos so now you can edit videos with more clarity of how to compose them, what works and what doesn't and we built a free tool for you based on this model yes, totally free see details below

AmirMušić

847,898 views • 3 months ago

Prince Harry: “It’s fantastic to see how HALO is using modern technologies to speed up the process of demining. The improvements and the ability to clear landmines faster is extraordinary.” The Duke of Sussex today joined HALO for a demonstration of the latest demining innovations in 🇺🇦 #Ukraine, where we’re using advanced technologies to clear land with greater efficiency and precision. This means we can make land safe more quickly, so families can return home, farmers can safely grow crops and children can play without fear. 🔗 Learn more:

Prince Harry: “It’s fantastic to see how HALO is using modern technologies to speed up the process of demining. The improvements and the ability to clear landmines faster is extraordinary.” The Duke of Sussex today joined HALO for a demonstration of the latest demining innovations in 🇺🇦 #Ukraine, where we’re using advanced technologies to clear land with greater efficiency and precision. This means we can make land safe more quickly, so families can return home, farmers can safely grow crops and children can play without fear. 🔗 Learn more:

The HALO Trust

44,755 views • 3 months ago

Teaser trailer to my short film “Je t’aime, Julie” based on the chronicles of Julie D’aubigny who is a bisexual swordswoman in the 17th century ⚔️ Directed by Me and my brother! We are doing a festival run at the moment and using this as a proof of concept for a feature length film! Thank you all so much for the support, I can’t wait to share this project with all of you ♥️ Here is our kickstarter with more information if you are interested:

Teaser trailer to my short film “Je t’aime, Julie” based on the chronicles of Julie D’aubigny who is a bisexual swordswoman in the 17th century ⚔️ Directed by Me and my brother! We are doing a festival run at the moment and using this as a proof of concept for a feature length film! Thank you all so much for the support, I can’t wait to share this project with all of you ♥️ Here is our kickstarter with more information if you are interested:

Fiona Nova

1,776,361 views • 2 years ago

Make Pixels Dance: High-Dynamic Video Generation paper page: Creating high-dynamic videos such as motion-rich actions and sophisticated visual effects poses a significant challenge in the field of artificial intelligence. Unfortunately, current state-of-the-art video generation methods, primarily focusing on text-to-video generation, tend to produce video clips with minimal motions despite maintaining high fidelity. We argue that relying solely on text instructions is insufficient and suboptimal for video generation. In this paper, we introduce PixelDance, a novel approach based on diffusion models that incorporates image instructions for both the first and last frames in conjunction with text instructions for video generation. Comprehensive experimental results demonstrate that PixelDance trained with public data exhibits significantly better proficiency in synthesizing videos with complex scenes and intricate motions, setting a new standard for video generation.

Make Pixels Dance: High-Dynamic Video Generation paper page: Creating high-dynamic videos such as motion-rich actions and sophisticated visual effects poses a significant challenge in the field of artificial intelligence. Unfortunately, current state-of-the-art video generation methods, primarily focusing on text-to-video generation, tend to produce video clips with minimal motions despite maintaining high fidelity. We argue that relying solely on text instructions is insufficient and suboptimal for video generation. In this paper, we introduce PixelDance, a novel approach based on diffusion models that incorporates image instructions for both the first and last frames in conjunction with text instructions for video generation. Comprehensive experimental results demonstrate that PixelDance trained with public data exhibits significantly better proficiency in synthesizing videos with complex scenes and intricate motions, setting a new standard for video generation.

AK

102,444 views • 2 years ago

🚀 Introducing HunyuanVideo-Avatar, a model jointly developed by Tencent Hunyuan and Tencent Music, bringing photos to life. ✅ Upload a photo + audio — auto-detect scene context & emotion, then generate lifelike speech/singing with dynamic visuals. ✅ Supports multi-style, multi-species scenarios and excels in multi-character interactions. ✅ Built for short video creation, e-commerce, advertising, and more — already deployed in multiple apps across Tencent Music Entertainment Group. The single-character mode of HunyuanVideo-Avatar is now open-sourced and live on the Hunyuan website, supporting audio up to 14 seconds for video generation. The multi-character mode will be open-sourced soon. Try it on: Project Page: Github: Tech report: #Tencent #Hunyuan

Tencent Hy

218,408 views • 1 year ago

🚀We are thrilled to open-source Hunyuan-GameCraft, a high-dynamic interactive game video generation framework built on HunyuanVideo. It generates playable and physically realistic videos from a single scene image and user action signals, empowering creators and developers to "direct" games with first-person or third-person perspectives. Key Advantages: 🔹High Dynamics: Unifies standard keyboard inputs into a shared continuous action space, enabling high-precision control over velocity and angle. This allows for the exploration of complex trajectories, overcoming the stiff, limited motion of traditional models. It can also generate dynamic environmental content like moving clouds, rain, snow, and water flow. 🔹Long-term Consistency: Uses hybrid history condition to preserve the original scene information after significant movement. 🔹Significant Cost Reduction: No need for expensive modeling/rendering. PCM distillation compresses inference steps, boosting speed and lowering costs. This allows the quantized 13B model to run on consumer-grade GPUs like the RTX 4090. Project Page: Code: Technical Report: Hugging Face：

🚀We are thrilled to open-source Hunyuan-GameCraft, a high-dynamic interactive game video generation framework built on HunyuanVideo. It generates playable and physically realistic videos from a single scene image and user action signals, empowering creators and developers to "direct" games with first-person or third-person perspectives. Key Advantages: 🔹High Dynamics: Unifies standard keyboard inputs into a shared continuous action space, enabling high-precision control over velocity and angle. This allows for the exploration of complex trajectories, overcoming the stiff, limited motion of traditional models. It can also generate dynamic environmental content like moving clouds, rain, snow, and water flow. 🔹Long-term Consistency: Uses hybrid history condition to preserve the original scene information after significant movement. 🔹Significant Cost Reduction: No need for expensive modeling/rendering. PCM distillation compresses inference steps, boosting speed and lowering costs. This allows the quantized 13B model to run on consumer-grade GPUs like the RTX 4090. Project Page: Code: Technical Report: Hugging Face：

Tencent Hy

169,638 views • 11 months ago

With news breaking that Israel and the U.S have begun operations against Iran, it is probably a good time to revisit the following clip from Episode 178 of Absolute Storm (2/25/26). We laid out a multitude of comms that ourselves and others picked up on during the SOTU that all seemed to point towards an activation code being given. We even said that the next 48 hours would be very important. Give it a watch to see how Al Green and the length of the speech - as well as other factors - led us to believe strikes on Iran were imminent.

With news breaking that Israel and the U.S have begun operations against Iran, it is probably a good time to revisit the following clip from Episode 178 of Absolute Storm (2/25/26). We laid out a multitude of comms that ourselves and others picked up on during the SOTU that all seemed to point towards an activation code being given. We even said that the next 48 hours would be very important. Give it a watch to see how Al Green and the length of the speech - as well as other factors - led us to believe strikes on Iran were imminent.

Paul Fleuret

13,305 views • 5 months ago

OK SO FINALLY I have a means to show you what I have been saying for a while.......Watch this clip, share it, so people understand what we are saying here!!! In this Clip, JLR© is having a discussion with Watching the World , and we get to see that front door lock!! Here is what I have been saying.....The Front door is a material called EXPANDED METAL, YOU CAN SEE THROUGH IT. The Door IS a Security Door, that you would have to cut through with two locks. A Lock that can be unlocked via a key on the handle, and a DEADbolt lock that would be locked on the inside of the door. This PREVENTS someone from being able to attack you when you OPEN the HOUSE door that is inside it, and behind this screen door. That door only opens when that deadbolt is unlocked from inside! ALso notice we see NO damage to the door or frame in this video!!

OK SO FINALLY I have a means to show you what I have been saying for a while.......Watch this clip, share it, so people understand what we are saying here!!! In this Clip, JLR© is having a discussion with Watching the World , and we get to see that front door lock!! Here is what I have been saying.....The Front door is a material called EXPANDED METAL, YOU CAN SEE THROUGH IT. The Door IS a Security Door, that you would have to cut through with two locks. A Lock that can be unlocked via a key on the handle, and a DEADbolt lock that would be locked on the inside of the door. This PREVENTS someone from being able to attack you when you OPEN the HOUSE door that is inside it, and behind this screen door. That door only opens when that deadbolt is unlocked from inside! ALso notice we see NO damage to the door or frame in this video!!

Capt KEVman!!!

46,092 views • 5 months ago

🚨 🚨 🇱🇧 🇺🇸 Secretary Rubio on Lebanon: - And so how do you defeat Hizballah? How do you replace Hizballah? You defeat and replace them with a government that is strong enough to be the sole force of arms in the country, basically with the security of Lebanon in the hands of the Lebanese Armed Forces, and a government that’s able to provide prosperity for its citizens. So we talked about the opportunities to do that. This is not just about miliary stuff. This is about how can we attract more U.S. and international investment into Lebanon. This is about things like how can we restart direct flights between the United States and Lebanon that are – right now, you can’t do that. So we’re going to start a process with them and see if that’s possible and when that can be done. -So we had a good meeting, and I think he’ll leave on Tuesday. He’ll have a very positive meeting with President Trump. And I think President Aoun will be able to return to Lebanon and say to the people of Lebanon that he is the one that’s getting positive results for the people

🚨 🚨 🇱🇧 🇺🇸 Secretary Rubio on Lebanon: - And so how do you defeat Hizballah? How do you replace Hizballah? You defeat and replace them with a government that is strong enough to be the sole force of arms in the country, basically with the security of Lebanon in the hands of the Lebanese Armed Forces, and a government that’s able to provide prosperity for its citizens. So we talked about the opportunities to do that. This is not just about miliary stuff. This is about how can we attract more U.S. and international investment into Lebanon. This is about things like how can we restart direct flights between the United States and Lebanon that are – right now, you can’t do that. So we’re going to start a process with them and see if that’s possible and when that can be done. -So we had a good meeting, and I think he’ll leave on Tuesday. He’ll have a very positive meeting with President Trump. And I think President Aoun will be able to return to Lebanon and say to the people of Lebanon that he is the one that’s getting positive results for the people

ToniMrad

27,695 views • 13 days ago

Elon Musk: Despite starting from scratch six months ago, Grok Imagine now generates more videos and images than everyone else combined. “My prediction is that most of AI compute is going to be real-time video understanding and real-time video generation. And we expect to be the leaders in that. It's worth emphasizing these points that six months ago we had basically nothing, very weak in video and image generation and editing and went in six months to the number one spot and in fact generating more videos and images than everyone else combined. We're going to do the same thing with coding and we're going to do the same thing with Macrohard and. I think people will be pretty impressed with the Grok 4.2 model that's coming out. It's a significant improvement. That's the small version of our new model. So we'll have a medium and a large version that are even more intelligent.” From: xAI All-Hands, February 10, 2026

Elon Musk: Despite starting from scratch six months ago, Grok Imagine now generates more videos and images than everyone else combined. “My prediction is that most of AI compute is going to be real-time video understanding and real-time video generation. And we expect to be the leaders in that. It's worth emphasizing these points that six months ago we had basically nothing, very weak in video and image generation and editing and went in six months to the number one spot and in fact generating more videos and images than everyone else combined. We're going to do the same thing with coding and we're going to do the same thing with Macrohard and. I think people will be pretty impressed with the Grok 4.2 model that's coming out. It's a significant improvement. That's the small version of our new model. So we'll have a medium and a large version that are even more intelligent.” From: xAI All-Hands, February 10, 2026

ELON CLIPS

12,007 views • 5 months ago

AI is opening St Peter’s Basilica to the world, including a new generation, in a language that is more accessible for the times we live in. What a privilege and honor to partner with the Vatican on this important project. You can visit here: yves ubelmann Vatican News

AI is opening St Peter’s Basilica to the world, including a new generation, in a language that is more accessible for the times we live in. What a privilege and honor to partner with the Vatican on this important project. You can visit here: yves ubelmann Vatican News

Brad Smith

15,122 views • 1 year ago

no one loves txt more than txt😭 🐱txt..is the the best case scenario of a group project. i think it’s a move that was close to a miracle 🐧 i know we are team that can go even further and we have so much more to show so i want to prove ourselves 🦊 i think we are a very experimental, adventurous, and fun group. i’ve always preferred something special and unique over things that are obvious or ordinary. i believe that is exactly the kind of group we are 🧸i hope this album becomes an opportunity for our moas to love this version of tomorrow x together, a team that has become even healthier and more mature 🐰 i want our team to do even better and fly even higher, and i feel like that’s going to happen. it's not like based on any specific fact but i just have a feeling. i think we’re going to be even more successful

no one loves txt more than txt😭 🐱txt..is the the best case scenario of a group project. i think it’s a move that was close to a miracle 🐧 i know we are team that can go even further and we have so much more to show so i want to prove ourselves 🦊 i think we are a very experimental, adventurous, and fun group. i’ve always preferred something special and unique over things that are obvious or ordinary. i believe that is exactly the kind of group we are 🧸i hope this album becomes an opportunity for our moas to love this version of tomorrow x together, a team that has become even healthier and more mature 🐰 i want our team to do even better and fly even higher, and i feel like that’s going to happen. it's not like based on any specific fact but i just have a feeling. i think we’re going to be even more successful

녱🐱

42,760 views • 3 months ago

Dear John, @akwakwaofrealestate Thank you for the update. We are aware that the contractor is still actively working on the site; Mobil - Ikota drainage construction. I understand that work had to slow down over the past few weeks due to increased water levels from recent heavy rainfall. This is a common challenge in coastal and water-based infrastructure projects, and we commend the contractor for prioritizing safety and site management under these conditions. Morso, the portion the contractors are working on now cannot be dammed hence the water backlog in some areas. While we apologize for this inconveniences experienced, we also implore residents to be patient and show understanding with the Government in the provision of the resilient infrastructure. Our team remains in close communication with the contractor to monitor progress and ensure the project stays on track. The commitment to completing this critical drainage infrastructure is unwavering, and we are confident that the contractor will resume full operations as soon as the water levels permit. We urge residents to remain patient as we work through these natural challenges. The completion of this project remains a top priority for the Lagos State Government. Thank you! TW ************* Project Brief LINING OF LEKKI-IKOTA DRAINAGE CHANNEL PROJECT SUMMARY 1. Total Length of Drain: 4.3km 2. Total Length Excavated: 2.75km 3. Stabilization: 1.89km 4. Total Length of Blinding Done: 1.43km 5. Total Length of Reinforcement: 1.43km 6. Total Length of Concrete Base Done: 1.32km. 7. Total Length Formwork Done: 1.43km 8. Total Length of Concrete Wall Done: 1.32km

Dear John, @akwakwaofrealestate Thank you for the update. We are aware that the contractor is still actively working on the site; Mobil - Ikota drainage construction. I understand that work had to slow down over the past few weeks due to increased water levels from recent heavy rainfall. This is a common challenge in coastal and water-based infrastructure projects, and we commend the contractor for prioritizing safety and site management under these conditions. Morso, the portion the contractors are working on now cannot be dammed hence the water backlog in some areas. While we apologize for this inconveniences experienced, we also implore residents to be patient and show understanding with the Government in the provision of the resilient infrastructure. Our team remains in close communication with the contractor to monitor progress and ensure the project stays on track. The commitment to completing this critical drainage infrastructure is unwavering, and we are confident that the contractor will resume full operations as soon as the water levels permit. We urge residents to remain patient as we work through these natural challenges. The completion of this project remains a top priority for the Lagos State Government. Thank you! TW ************* Project Brief LINING OF LEKKI-IKOTA DRAINAGE CHANNEL PROJECT SUMMARY 1. Total Length of Drain: 4.3km 2. Total Length Excavated: 2.75km 3. Stabilization: 1.89km 4. Total Length of Blinding Done: 1.43km 5. Total Length of Reinforcement: 1.43km 6. Total Length of Concrete Base Done: 1.32km. 7. Total Length Formwork Done: 1.43km 8. Total Length of Concrete Wall Done: 1.32km

Tokunbo Wahab

60,524 views • 10 months ago

We are so happy to share that our first project for GraceOaey was a big success! Thank you so much to the PuenRaks who gave their support and donations. Your kindness made this project possible, and we are very grateful. To GraceOaey, we love and support you always. And we will continue to cheer for both of you in everything. Merry Christmas and Happy New Year! This is just the beginning! Let’s keep working together to make more projects happen for GraceOaey. With love, GRACEOAEY OFFICIAL PH #MateTheSeries #เกรซโอ๋เอ๋ #GraceOaey #GenlongAoey #PuenRakFandom #Purrner #TeeRakLak #Gracebudsarin #Oaeyponchanok

We are so happy to share that our first project for GraceOaey was a big success! Thank you so much to the PuenRaks who gave their support and donations. Your kindness made this project possible, and we are very grateful. To GraceOaey, we love and support you always. And we will continue to cheer for both of you in everything. Merry Christmas and Happy New Year! This is just the beginning! Let’s keep working together to make more projects happen for GraceOaey. With love, GRACEOAEY OFFICIAL PH #MateTheSeries #เกรซโอ๋เอ๋ #GraceOaey #GenlongAoey #PuenRakFandom #Purrner #TeeRakLak #Gracebudsarin #Oaeyponchanok

GRACEOAEY OFFICIAL PH (CLOSED)

123,991 views • 1 year ago

Microsoft unveiled an AI badge with a camera that can record tasks and track your entire workday. "Look, it's good and bad. For example, we sometimes record meetings so we can have a transcript. Then we put the transcript through AI, and then we get action items, we get all the notes, and we get everything. So having a constant transcript of everything that is going on is very useful if you're thinking about productivity." "And then you have the privacy concerns. Big corporations, they wanna get more data on how and what their employees are doing to train models to probably replace them down the line. So I'm very split on this one."

Microsoft unveiled an AI badge with a camera that can record tasks and track your entire workday. "Look, it's good and bad. For example, we sometimes record meetings so we can have a transcript. Then we put the transcript through AI, and then we get action items, we get all the notes, and we get everything. So having a constant transcript of everything that is going on is very useful if you're thinking about productivity." "And then you have the privacy concerns. Big corporations, they wanna get more data on how and what their employees are doing to train models to probably replace them down the line. So I'm very split on this one."

PBD Podcast

27,975 views • 1 month ago

Currently testing Hunyuan brand new 2.5 model for image to 3D. YUP! It's MAJOR improvement over 2.0, especially with making details look sharper, it can also reconstruct more detail. It also generates PBR materials now and you can automatically rig the character on the website. It's much better than Trellis and Tripo from my testing! Took it into Blender as well to compare Hunyuan 2.5 (on the left) and Hunyuan 2.0 (on the right). I wanted to see the mesh with a solid and material preview. If you're doing 3D printing, these would probably work very well. However...the biggest issue here is still topology which is a complete and utter mess, there's also still issues with texture projection...and the entire model is a single mesh, there needs to be a method to generate the model in parts BUT this tech is certainly improving. As someone that is also a 3D artist, I actually want to see what this tech is capable of generating to output 3D models.

Travis Davids

45,357 views • 1 year ago