Video wird geladen...

Video konnte nicht geladen werden

Zur Startseite

Quality of life updates just released for Node Banana - image grid splitting: should work with any grid layout - resizing improvements when you have multiple nodes selected - image history, so you don't lose images when regenerating.

ReflctWillie's profile picture

willie

6,288 subscribers

39,339 Aufrufe • vor 8 Monaten •via X (Twitter)

LIVE

Anya Rossi• Live Now

Private livecam show

0 Kommentare

Keine Kommentare verfügbar

Kommentare vom Original-Post werden hier angezeigt

Ähnliche Videos

Quality of life improvements for Node Banana: - Grid splitting node, it's borderline sexual. Now grid splitting can happen during the workflow, with preset node settings. - MacOS trackpad is nice to use now - Styling updates so you know what node is selected
0:44

Quality of life improvements for Node Banana: - Grid splitting node, it's borderline sexual. Now grid splitting can happen during the workflow, with preset node settings. - MacOS trackpad is nice to use now - Styling updates so you know what node is selected

willie

14,857 Aufrufe • vor 8 Monaten

Found a way to make nano-banana-pro insanely cost-effective on fal I built a workflow that generates 9 4K images for a total of $0.21. That’s just $0.023 per image. That is a ~92% cost reduction. The key is grid generation. I added a 'Crop Node' as a workflow utility specifically for this. It takes the grid output, splits it, and then we upscale everything using SeedVR
0:20

Found a way to make nano-banana-pro insanely cost-effective on fal I built a workflow that generates 9 4K images for a total of $0.21. That’s just $0.023 per image. That is a ~92% cost reduction. The key is grid generation. I added a 'Crop Node' as a workflow utility specifically for this. It takes the grid output, splits it, and then we upscale everything using SeedVR

ilker

148,418 Aufrufe • vor 8 Monaten

New Framer component drop: Grid Warp ⚡️ Interactive, grid-based image distortion with smooth release + chromatic aberration. And of course, no code needed (that was my job). Comment “WARP” and I’ll send it to you for free. P.S. Follow me so I can DM you.
0:46

New Framer component drop: Grid Warp ⚡️ Interactive, grid-based image distortion with smooth release + chromatic aberration. And of course, no code needed (that was my job). Comment “WARP” and I’ll send it to you for free. P.S. Follow me so I can DM you.

David McBacon

13,263 Aufrufe • vor 1 Jahr

Adobe just released its first “commercially safe” Adobe Firefly Video in public beta. You can access it in Adobe Premiere & a new Firefly mobile app. Scene to Image & Text to Image are interesting for 3D artists, as you have a lot of control over your images. What do you think?
0:56

Adobe just released its first “commercially safe” Adobe Firefly Video in public beta. You can access it in Adobe Premiere & a new Firefly mobile app. Scene to Image & Text to Image are interesting for 3D artists, as you have a lot of control over your images. What do you think?

E.J. Hassenfratz

22,897 Aufrufe • vor 1 Jahr

This update cost me my soul :0 You can now download the update link With better "Quick Edit" for #Blender. You don't have to rely on just one image. Crop what you want to edit. No ugly outline for images.
0:32

This update cost me my soul :0 You can now download the update link With better "Quick Edit" for #Blender. You don't have to rely on just one image. Crop what you want to edit. No ugly outline for images.

Hirourk

41,590 Aufrufe • vor 2 Jahren

We just released Photoshop AI Assistant 🎉 I edited my images with text keeping consistency. It understands context of the image, so can select particular parts of the image you want to edit. And, of course, it can name your layers 😅 and do many other things. Enjoy!
0:33

We just released Photoshop AI Assistant 🎉 I edited my images with text keeping consistency. It understands context of the image, so can select particular parts of the image you want to edit. And, of course, it can name your layers 😅 and do many other things. Enjoy!

Kris Kashtanova

18,169 Aufrufe • vor 2 Monaten

Just when you thought the climate scam couldn’t get any worse. Instead of getting a rebate now Aussies will be taxed for sending their solar energy to the grid. Reprehensible.
0:54

Just when you thought the climate scam couldn’t get any worse. Instead of getting a rebate now Aussies will be taxed for sending their solar energy to the grid. Reprehensible.

Kat A 🌸

208,134 Aufrufe • vor 1 Jahr

JAMES CARVILLE; "When you win, [the old image] goes away. It sounds so absurd to say this, but people have to be reminded: In politics, winning is everything. If you lose, your own people don’t like you. And if the Democrats have a good 2026, the image of the party will pop up."
1:19

JAMES CARVILLE; "When you win, [the old image] goes away. It sounds so absurd to say this, but people have to be reminded: In politics, winning is everything. If you lose, your own people don’t like you. And if the Democrats have a good 2026, the image of the party will pop up."

Bill Kristol

91,097 Aufrufe • vor 11 Monaten

All these demo videos make HEAD SWAPPING with Nano Banana look so easy, but then you give it a try and you're like... uh... what? Why didn't that work? Here's what I've found. Nano Banana reads your image, almost literally, so if you write on the image, it reads the text. This is how Higgsfield AI 🧩 has capitalized on the tech: "Write on the image" and give it direction, right? Totally true, but you don't need Higgi to write on your image. Nano Banana will understand your direction regardless of where you write on your image. On one hand, Higgi is really smart, because they're hranessing the tech in a unique way, but the whole "Higgsfield's Banana Placement" is a bit of a misnomer. It's more of a "Banana Placement" and Higgi is just giving you a sort of basic Photoshop-type tool to work with (again, pretty smart), but the real tech is the Banana. 🍌 This is how I head swapped heads in Runway, but Nano Banana maintains the aesthetic qualities of your image almost perfectly, whereas Runway Reference spits out a very Gen-4 looking image. I like using Nano in Freepik (now Magnific), mainly because it's fast and I can get 4 gens at a time, and you need to gen a dozen times of so before you get a winner (most of the time). I was pumped when I saw Freepik introduce the @ reference feature, just like Runway has, but it doesn't seem to work for head swapping. My guess is because that's not really how Nano Banana tech works... ideally. Marco is the person I saw using this "A" and "B" method, back when Nano was on LM Arena, and man-oh-man, it just works... like a charm. You need experiment with how much of the face you blot out, and the angle and facial expression of your new head if you want the blend to be perfect. All of the results in this video are 100% Nano Banana. I did not do any Photoshop work to the images after the fact. I really hope this helps. Let me know if you have any questions. I'm happy to help. And I'll keep posting videos like this if you guys find them useful. Let me know! And if you want more serious, one-on-one AI consultation you can throw something on the books here:
3:05

All these demo videos make HEAD SWAPPING with Nano Banana look so easy, but then you give it a try and you're like... uh... what? Why didn't that work? Here's what I've found. Nano Banana reads your image, almost literally, so if you write on the image, it reads the text. This is how Higgsfield AI 🧩 has capitalized on the tech: "Write on the image" and give it direction, right? Totally true, but you don't need Higgi to write on your image. Nano Banana will understand your direction regardless of where you write on your image. On one hand, Higgi is really smart, because they're hranessing the tech in a unique way, but the whole "Higgsfield's Banana Placement" is a bit of a misnomer. It's more of a "Banana Placement" and Higgi is just giving you a sort of basic Photoshop-type tool to work with (again, pretty smart), but the real tech is the Banana. 🍌 This is how I head swapped heads in Runway, but Nano Banana maintains the aesthetic qualities of your image almost perfectly, whereas Runway Reference spits out a very Gen-4 looking image. I like using Nano in Freepik (now Magnific), mainly because it's fast and I can get 4 gens at a time, and you need to gen a dozen times of so before you get a winner (most of the time). I was pumped when I saw Freepik introduce the @ reference feature, just like Runway has, but it doesn't seem to work for head swapping. My guess is because that's not really how Nano Banana tech works... ideally. Marco is the person I saw using this "A" and "B" method, back when Nano was on LM Arena, and man-oh-man, it just works... like a charm. You need experiment with how much of the face you blot out, and the angle and facial expression of your new head if you want the blend to be perfect. All of the results in this video are 100% Nano Banana. I did not do any Photoshop work to the images after the fact. I really hope this helps. Let me know if you have any questions. I'm happy to help. And I'll keep posting videos like this if you guys find them useful. Let me know! And if you want more serious, one-on-one AI consultation you can throw something on the books here:

Jordan Daniel Chesney

62,089 Aufrufe • vor 11 Monaten

Energy Secretary Chris Wright: "With wind and solar, you don't know when they're gonna be there and when they're gonna go away. That's just not productive for the electricity grid. They end up just making the reliable stuff have to turn down when the wind is blowing, turn back up again when the wind dies."
1:35

Energy Secretary Chris Wright: "With wind and solar, you don't know when they're gonna be there and when they're gonna go away. That's just not productive for the electricity grid. They end up just making the reliable stuff have to turn down when the wind is blowing, turn back up again when the wind dies."

Aaron Rupar

301,092 Aufrufe • vor 1 Jahr

fruit make up life hack for when you're in a pinch and don't have any make up with you 🤩🤩
0:33

fruit make up life hack for when you're in a pinch and don't have any make up with you 🤩🤩

Ahib Jane

258,236 Aufrufe • vor 8 Monaten

500,000 USDT earnings will now be shared to Node holders on a rolling basis. Leaderboard has been updated to reflect your wallet based earnings. In the coming weeks, you will be able to see the transaction hash next to it, once the funds have been sent. Rolling basis means, you receive 250$ once you hit 5 nodes, you receive another 1750$ when you hit 20 nodes (or you get directly 2000$ if you are at 20 nodes) and so on upto receiving a total of 25,000$ when you hit 100 Nodes. May the multiplier be with you. Node operations start shortly.
0:22

500,000 USDT earnings will now be shared to Node holders on a rolling basis. Leaderboard has been updated to reflect your wallet based earnings. In the coming weeks, you will be able to see the transaction hash next to it, once the funds have been sent. Rolling basis means, you receive 250$ once you hit 5 nodes, you receive another 1750$ when you hit 20 nodes (or you get directly 2000$ if you are at 20 nodes) and so on upto receiving a total of 25,000$ when you hit 100 Nodes. May the multiplier be with you. Node operations start shortly.

GPUNET

22,646 Aufrufe • vor 2 Jahren

We've shipped several quality-of-life improvements to Cursor 3. They bring a little more delight when you are orchestrating agents. Just like in your terminal, you can now split agents for multi-tasking in Cursor.
0:20

We've shipped several quality-of-life improvements to Cursor 3. They bring a little more delight when you are orchestrating agents. Just like in your terminal, you can now split agents for multi-tasking in Cursor.

Cursor

370,203 Aufrufe • vor 4 Monaten

SimInvest V Interview Q10: When V is with best friend, when do you feel most free? 🐯 Firstly, it’s great that I can show myself without putting on any kind of image or act a certain way. Because, when on stage or in photoshoots, commercials, or video shoots, rather than a person called “Kim Taehyung”, I have to become "V of BTS." Without realising it, I’ve just naturally created this image. But when I'm with friends, I can just be myself and have fun. 😄
0:43

SimInvest V Interview Q10: When V is with best friend, when do you feel most free? 🐯 Firstly, it’s great that I can show myself without putting on any kind of image or act a certain way. Because, when on stage or in photoshoots, commercials, or video shoots, rather than a person called “Kim Taehyung”, I have to become "V of BTS." Without realising it, I’ve just naturally created this image. But when I'm with friends, I can just be myself and have fun. 😄

Taehyung Naver

21,419 Aufrufe • vor 1 Jahr

We all know adding media automatically boosts your posts on the algorithm So I built a tool that can generate any image with your tweets using Nano Banana. You can search for the right memes, create infographics and more with JUST text
3:58

We all know adding media automatically boosts your posts on the algorithm So I built a tool that can generate any image with your tweets using Nano Banana. You can search for the right memes, create infographics and more with JUST text

Andros

56,077 Aufrufe • vor 11 Monaten

Nano-Banana moment for Video models soon? In an interview two weeks ago, Demis Hassabis talked about the future of image and video models "When Video converges with Language models you will see some interesting combinaton of Capabilites"
0:31

Nano-Banana moment for Video models soon? In an interview two weeks ago, Demis Hassabis talked about the future of image and video models "When Video converges with Language models you will see some interesting combinaton of Capabilites"

AiBattle

47,820 Aufrufe • vor 8 Monaten

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 7 Monaten

This is the Existence Grid. It was inspired by the idea of seeing your entire life laid out in weeks. How much you have created and how much you still have to create. The grid is designed to show your life in years, months, weeks, and days. Not just as time passing, but as time created. For me, that is 51 years, 618 months, 2,686 weeks, and 18,802 days. It all began with one time block. June 28th, 1974 at 6:50am. The first block of my existence. This grid becomes the record of what you have created. The true story of you. Learn more about Existence here:
1:53

This is the Existence Grid. It was inspired by the idea of seeing your entire life laid out in weeks. How much you have created and how much you still have to create. The grid is designed to show your life in years, months, weeks, and days. Not just as time passing, but as time created. For me, that is 51 years, 618 months, 2,686 weeks, and 18,802 days. It all began with one time block. June 28th, 1974 at 6:50am. The first block of my existence. This grid becomes the record of what you have created. The true story of you. Learn more about Existence here:

Rob Dyrdek

99,739 Aufrufe • vor 8 Monaten