Share this page!

Enter URL or ID to Unroll

You can paste full URL like: https://x.com/threadreaderapp/status/1644127596119195649
or just the ID like: 1644127596119195649

How to get URL link on X (Twitter) App

  1. On the Twitter thread, click on or icon on the bottom
  2. Click again on or Share Via icon
  3. Click on Copy Link to Tweet
  4. Paste it above and click "Unroll Thread"!
  5. More info at Twitter Help
TechHalla Profile picture
@techhalla
May 21 • 21 tweets • 6 min read • Read on X
Less than 24 hours since Google dropped Veo3 and people are already creating wild stuff!

13 insane examples 🧵👇

1. A giraffe riding a bike in NYC
2. A professor doing class on Gen Z
3. Car show
4. ASMR channel
5. Car crash
6. Un café con leche, por favor
7. 100 men vs a gorilla at a rave
8. A horse walks into a bar
9. Stunning physics
10. A man rapping about Veo3
11. Mashala chai
12. Mage interview
13. Pythagoras explaining his theorem
Hope you liked this roundup!

I’ll be testing my prompts on Veo3 very soon, so stay tuned to @techhalla and don’t forget to drop a like, repost, and bookmark on the main post, it really helps me out!
Another big compilation 👇🏻
And don’t forget to check out the hottest AI event happening right now @upscaleconf in SFO!
@upscaleconf If you’re looking to try an amazing video model at an affordable price… @higgsfield_ai is the way to go!
@upscaleconf @higgsfield_ai And @minchoi recap, always a must
@upscaleconf @higgsfield_ai @minchoi A perfect pharmaceutical commercial 👇🏻
@upscaleconf @higgsfield_ai @minchoi They can talk! 🤯
@upscaleconf @higgsfield_ai @minchoi You can also check out my own roundup after testing it out on Freepik with early access

• • •

Missing some Tweet in this thread? You can try to force a refresh
　

Keep Current with TechHalla

TechHalla Profile picture

Stay in touch and get notified when new unrolls are available from this author!

Read all threads

This Thread may be Removed Anytime!

PDF

Twitter may remove this content at anytime! Save it as PDF for later use!

Try unrolling a thread yourself!

how to unroll video
  1. Follow @ThreadReaderApp to mention us!

  2. From a Twitter thread mention us with a keyword "unroll"
@threadreaderapp unroll

Practice here first or read more on our help page!

More from @techhalla

TechHalla Profile picture
Aug 26
nano-banana is now available and unlimited! 🍌

Freepik just added it to their suite along with Veo 3 and Kling 2.1... and it’s the perfect combo!

this video is just a glimpse of what you can do now. Prompts, tips, and all the info right here 👇
I don't know how they pulled it off, but @freepik just added the most wanted image model right now: nano-banana by Google.

Yep, the one that lets you edit, compose, get those crazy angles... You can use it unlimited now!

Let me show you how it works 👇 Image
Let’s start with creating the base images. I used Imagen 4 Ultra, which is unlimited on Freepik (I’ll tell you how in a sec).

We’ll move on to nano-banana next, but first, I’ve dropped the prompts in the ALTs. A cinematic, moody close-up of a desert viper in sharp focus, its dusty scales glinting under the low desert sun | shallow depth of field | in the background, a heavily blurred, rundown redneck RV trailer with rusted panels and clutter around it | hazy air and gritty texture | 85mm film lens effect | golden hour lighting | analog film grain | muted ochre and beige tones
A cinematic medium shot of a chubby, balding man with a skullet haircut and thick mustache that drops down the corners of his mouth, wearing aviator sunglasses | he's holding a phone to his ear with a serious, badass expression | inside a filthy, rundown RV with torn upholstery, yellowed curtains, and cluttered surfaces | harsh desert sunlight seeping through dusty windows | gritty, moody lighting | analog film style | 35mm frame
A cinematic wide shot of two rugged workers loading a heavy wooden crate full of ripe bananas into the back of an old, dusty truck | warm, golden hour lighting casting long shadows | the truck's tailgate is open, showing a worn metal floor and scattered debris | the workers are sweaty, in rolled-up sleeves and work pants | desert environment with dry ground and faded vegetation in the distance | shot on 35mm film with high contrast and rich color grading --style raw
A hyper-realistic photograph of a custom redneck-style quad bike designed for the desert | dusty, oversized off-road tires, rusted metal body with DIY modifications like duct tape, rope, beer can holder, and mismatched paint | worn-out leather seat and exposed engine parts | shot on white background in studio lighting | high-definition detail with subtle shadows under the quad | product photography style
Read 10 tweets
TechHalla Profile picture
Aug 23
AI lets you express your emotions in ways you never could’ve even imagined.

In this thread, I’ll show you how I created this video, including the prompts and styles I used on freepik 👇
The @freepik team added Kling 2.1 initial/final frame to their suite, unlocking all kinds of wild transitions in any style.

To make it work best, you need strong stills.

I used the Mystic 2.5 model with the Fine Art style, prompts are in the screenshots. Image
Image
Image
Image
Now let’s jump to video generation. Choose the Kling 2.1 model in 1080p, upload the two frames you want to animate, and write a detailed prompt.

Make sure you include everything you want to happen. A JSON prompt works great for that.

Example’s in the ALT.
{ "scene_description": "A surreal transformation where a woman approaches a giant eye surrounded by plants; as she walks forward, the eye closes, the surrounding plants decay into dust, her appearance changes, and dense foam fills the scene, revealing a new environment with a different giant eye in a desert cave.", "visual_style": "Hyper-realistic surrealism with cinematic polish, blending organic and fantastical elements.", "camera_movement": "Slow, steady dolly-in toward the woman from behind, subtly tracking her movement as the...
Read 6 tweets
TechHalla Profile picture
Aug 23
Veo 3 is free for everyone this weekend.

So you’ll need good tips and prompts… and here I’m giving you more than 100 🧵👇🏻
1. Chimpanzees doing things
2. Insane ads
Read 18 tweets
TechHalla Profile picture
Aug 22
Been playing with Kling’s latest and I’m obsessed.

version 2.1 now includes start and end frames, and that changes the game again!

everything you need to know, prompts included, is in this thread 🧵👇
The @Kling_ai team has cooked up something huge.

We’ve been waiting for this for a while, and now we can finally use start and end frames for our videos.

That means total control and super polished results, with no morphing and way smoother motion. Let’s see it in action 👇 Image
First step before animating: pick stills that match what you want.

The model handles a lot, but it works best with similar colors, elements, or characters. That really helps, especially for pro projects where consistency matters.

Here’s an example and prompt’s in the ALT.
Image
Image
A brutal, towering orc warrior clad in filthy, blood-crusted fur and spiked iron armor, his monstrous muscles flexing as he stomps through the battlefield, dragging a massive gore-soaked axe, thick dreadlocks swinging violently, tusks jutting from a twisted snarl, pure rage in his eyes as he marches forward; he suddenly halts, staring with savage intensity straight ahead as if locking eyes with his enemy; the camera whips around him in a rapid, cinematic arc, then blasts forward across the battlefield at breakneck speed, flying low over mud, corpses, shattered weapons and burning debris, sl...
Read 9 tweets
TechHalla Profile picture
Aug 20
I’m sure you’ve seen tons of videos like this blowing up the internet...

but maybe you don’t realize how easy they are to make. All you need is Montra and the right prompts!

I’ll break it all down for you in this thread 🧵👇
The @Montra team built a platform where you can create full end to end videos just by interacting with their chat, Montra Director.

Check this out, we’ll start by creating our character, the famous Yeti!

Prompt's in the ALT. A realistic handheld selfie of the Abominable Snowman (Yeti) taken with an iPhone | front-facing camera angle | visible arm holding the phone | friendly, gentle expression | rugged fur covered in patches of snow | slightly distorted lens perspective | snowy Himalayan peaks in the blurry background | natural outdoor lighting | cold, pale blue color tones | motion blur on the edges | raw phone camera aesthetic
Image
We want to keep the same character, so our reference Yeti will always be the one we just made.

Then we pick the image and tell Montra’s Director what changes we want.

Now we’ll ask it to place our Yeti in different settings.

Prompts are in the ALTs. swap the background by a snowy village. keep the lighting the same. daylight
keep the yeti the same.swap the background by the interior of a bus. snowy village through its windows. keep the lighting the same. daylight
keep the yeti the same, showing a beer glass on one hand, the other holding the camera like a selfie.swap the background by a swimming pool party, luxury, women. keep the lighting the same. daylight
Image
Read 6 tweets
TechHalla Profile picture
Aug 19
lot of people insist that you can’t get consistent characters with AI.

don’t be one of them and follow the step-by-step tutorial on this thread 🧵👇

made with Multi-Ref by LTX Studio
Alright, first step is creating the character.

I took a real photo of myself, uploaded it to @LTXStudio, and with a super simple prompt:

make a 2D disney illustration version of this man. against white background

I turned it into a 2D illustration.Image
Image
Image
Now let’s generate some backgrounds.

This time, without using any reference images, I created these ones.

You’ll find the prompts in the ALTs. first-person POV from a snorricam mounted on a rider’s chest, looking down at the empty front seats of a roller coaster wagon | visible safety bar in front, detailed metallic structure, padded red seats | early morning light with soft shadows over the tracks | blurred background of roller coaster loops and steel supports | realistic perspective with slight lens distortion | cinematic depth of field, immersive theme park photography
a cinematic close-up of a cactus in the desert, with sharp focus on the spines and blurred background | depth of field effect emphasizing texture and form | golden hour lighting casting warm highlights and soft shadows on the cactus surface | dry sand and scattered rocks in the background | captured with a macro lens, ultra-realistic detail and shallow focus | nature photography with a filmic mood
a photo of a frosty beer mug resting on the wooden bar of an Irish tavern | background shows a warmly lit pub interior with dark wood furnishings, a television playing in the distance, and blurred patrons at tables | professional composition | shallow depth of field | cozy ambiance | captured with a Nikon D850 and 50mm lens | natural grain | cinematic lighting
aerial drone photo of a modern city skyline partially covered by soft, low-hanging clouds | tall skyscrapers piercing through the mist, morning light casting golden reflections on the glass buildings | dramatic atmosphere, city grid barely visible beneath the fog | cinematic, high-altitude perspective | captured with a DJI drone in high resolution | atmospheric realism, soft tones blending with architecture
Read 7 tweets

Did Thread Reader help you today?

Support us! We are indie developers!

This site is made by just two indie developers on a laptop doing marketing, support and development! Read more about the story.

Become a Premium Member ($3/month or $30/year) and get exclusive features!

Become Premium

Don't want to be a Premium member but still want to support us?

Make a small donation by buying us coffee ($5) or help with server cost ($10)

Donate via Paypal

Or Donate anonymously using crypto!

Ethereum

0xfe58350B80634f60Fa6Dc149a72b4DFbc17D341E copy

Bitcoin

3ATGMxNzCUFzxpMCHL5sWSt4DVtS8UqXpi copy

Thank you for your support!

Follow Us!

Send Email!

:(