Creative Director @PrimeVideo | Generative Technologist @watchonwonder House of David S1 & S2 | AI Educator & Consultant
- Tweets 131,450
- Following 710
- Followers 54,598
- Likes 176,057
Pinned Tweet
ChatGPT made a storyboard from a single reference image, then I used Seedance 2.0 to animate the storyboard shots:
130
318
3,487
305,710
🎬 Video Model Comparison: Text to video
• Google Veo 2
• OpenAI Sora
• Hunyuan Video
• Pika 2.1
• Alibaba Wanx 2.1
• Runway Gen-3
• Kling 1.6
• Luma Ray2
• Hailuo T2V-01
I used the same prompt on each model for 2 generations and chose the best results below.
Prompt: A majestic white tiger prowling silently through a dense jungle, sunlight piercing through the foliage to illuminate its fur; shot with a telephoto lens for depth, using soft natural lighting and creating a mysterious yet tranquil atmosphere.
Prompt credit: @LudovicCreator
Overall, the results are impressive for each one. We have many choices for AI video!
I could have tested more models, but this visual is crowded enough. Drop your results in the comments 👇
109
501
2,776
255,316
Unbelievable.
LTX Studio just launched their most advanced video model yet including:
→ Camera motion control
→ Keyframe animation
→ 30x faster renders
→ Open Source
Here's what it means for you:
67
192
2,299
80,663
Simple image prompts can produce beautiful results.
Prompt: "Stylized painting of a mountain range at sunrise"
10 examples from different image generators:
1. Lexica
87
299
2,205
710,300
🔹STORM🔹
Stanford University’s free, public app automates comprehensive research and report generation using AI to create Wikipedia-style articles with citations from web sources.
STORM is incredible.
I asked it to write a research paper about the use of AI in DNA analysis and predicting genetic traits - got a comprehensive article with citations.
It takes about 3 minutes and you can download the PDF:
34
247
1,661
176,268
11 new ChatGPT ideas from creators that will save you time and maximize your creativity:
57
275
1,391
225,198
Midjourney V6 understands natural language better than previous versions.
But we have a new way to tell stories -
Instead of a single scene, we can now present a narrative across multiple panels, each capturing a distinct moment in time.
I'll show you how:
48
157
1,216
197,011
Microsoft's Bing Image Creator is available.
Here's 10 styles I generated immediately:
58
143
1,226
361,730
I just found @napkin_ai and it is incredible!
Create visuals from any text instantly - just highlight the text and click the blue Spark button.
The Pro plan is FREE during the beta - link below:
44
137
1,123
97,309
⚡AI Video Model Comparison: Text to Video
9 Models included:
• Google Veo 2
• Adobe Firefly Video
• Hunyuan Video
• Pika 2.1 Turbo
• Genmo Mochi 1
• Runway Gen-3
• Kling 1.6
• Luma Ray2
•.Hailuo T2V-01 Director
I used the same prompt on each model x 2 generations and chose my favorite results.
Prompt: [Tilt up, Left circling] From the base of an ancient pyramid as storm clouds gather, lightning illuminating hieroglyphics. [Zoom out] Sand swirls in the wind. Epic scale, dramatic lighting.
Overall, the results are interesting and dramatic. I don't have a favorite, but you might!
62
230
1,096
80,225
I've been experimenting with DALL•E 3 in ChatGPT, compared to DALL•E 3 in Bing, and Midjourney.
Here are my results from a Game of Thrones-style map of medieval Europe. (Prompt in ALT)
DALL•E 3 in ChatGPT:
DALL•E 3 in Bing:
Midjourney:
ALT ChatGPT: Hyper-realistic 3D side view depiction of a 13th century European map, channelling the artistic style of the Game of Thrones intro. The elevated representation of kingdoms and nation-states provides a comprehensive and immersive view of the continent during that era.
ALT Bing: Hyper-realistic 3D side view depiction of a 13th century European map, channelling the artistic style of the Game of Thrones intro. The elevated representation of kingdoms and nation-states provides a comprehensive and immersive view of the continent during that era.
ALT Midjourney: Hyper-realistic 3D side view depiction of a 13th century European map, channelling the artistic style of the Game of Thrones intro. The elevated representation of kingdoms and nation-states provides a comprehensive and immersive view of the continent during that era.
60
113
969
270,803
I'm so excited!
Starting a new Creative Director role with Amazon Prime Video & Studio CoreTech 🎉
I get to continue working with current and emerging generative models to build efficient creative workflows.
354
20
1,014
37,170
🔳 Video Model Comparison: Image to Video
• Ray3
• Veo 3
• Seedance Pro
• Kling 2.1
• Midjourney Video
• Hailuo 02
The first video I posted is glitchy halfway through the individual model examples, so you can see the performance side-by-side below.
I was impressed by each of these:
🔳 Video Model Comparison: Image to Video
I was able to test @LumaLabsAI Ray3 early so I added it to the list below:
→ Ray3
→ Veo 3
→ Seedance Pro
→ Kling 2.1
→ Midjourney Video
→ Hailuo 02
I used the same prompt on each model x 2-4 times and chose my favorites:
A sleek white Audi R8 glides through a bustling city street at golden hour, camera tracking low from the front-left angle as reflections ripple across the car’s glossy surface. Buildings blur into streaks of glass and stone, emphasizing speed and elegance. The car roars forward with cinematic intensity, tires gripping the asphalt as the city rushes past.
60
125
966
135,834
Video Model Comparison: Image to video
I updated an older comparison with a new image and prompt to test in several models:
• Pika 2.2
• PixVerse 4
• Runway Gen-4
• Google Veo 3
• OpenAI Sora
• Kling 2.1 Pro
• Luma Ray2
• Hailuo I2V
I used the same prompt and Midjourney image in each model 2-3 times and chose my favorite results below:
63
137
857
98,414
🌊 Video Model Comparison: Image to Video
→ Google Veo 3
→ Moonvalley Marey
→ Hailuo MiniMax 02
→ Kling 2.1
→ Midjourney V1
→ Seedance Pro
Image: Midjourney
I ran the same prompt on each model 2-4 times and chose my favorite results below:
51
85
799
67,858
I can FINALLY share what I've been working on for the past 6 months!
I worked on the # 2 series streaming now on Amazon Prime: House of David
I joined the @watchonwonder post-production team to incorporate generative models in the pipeline:
166
70
761
63,532
✨AI Video Model Comparison: Image to Video
• Pika 2.0
• Sora
• Gen-3
• Kling 1.6
• Luma
• Hailuo
Image: Mystic 2.5 on @Magnific_AI
I used the same prompt on each model for 2-4 generations and picked my favorite results.
Take a moment to watch the individual videos following the comparison grid - don't dismiss a result because of the lack of movement, without looking at the details, such as:
Consistent color
Quality of stone pathway
Wind blowing grass & flowers
Waves moving realistically below
Lighthouse light flickering / moving
Full workflow below:
57
118
726
69,609
The era of lipdub AI is over.
Captions just launched Mirage.
The most advanced AI video model we’ve seen.
Here’s why:
58
102
722
118,108
Words cannot express how thrilled I am at seeing my images in Midjourney Magazine!
Thank you 💙
109
37
671
42,692
💄AI Video Model Comparison: Text to Video
10 Models included:
• Google Veo2
• Adobe Firefly Video
• Hunyuan Video
• Pika 2.1
• Wan-2.1
• Runway Gen-3
• Kling AI 1.6
• Luma Ray2
• Hailuo MiniMax
• Sora (not on grid, but in the full video)
I used the same prompt on each model x 4 generations and chose my favorite results below.
⭕ Prompt: A person stands in a bright, airy room, bathed in soft natural light streaming through a large window adorned with sheer white curtains. As they gaze into a shimmering mirror, their hand, elegantly manicured with a deep crimson polish, lifts a glossy lip applicator towards their lips. Their outfit, a flowing pastel blouse paired with chic high-waisted trousers, exudes effortless style, complemented by delicate gold jewelry that sparkles subtly in the light.
Applying lipstick, lip gloss, or any makeup seems to be extremely difficult for AI video models. It was hard to get a usable result for many of these.
This prompt was generated on @nimvideo with the Video Describe feature, and it worked better than my original prompt.
Let me know what you think of the results!
32
129
695
61,607
Nano Banana Image Prompt Workflow 🍌
Grid theme:
Prompt 1:
8-panel grid layout with clear grid lines separating each square:
Panel 1) sleek chrome samurai helmet centered on pure white background,
Panel 2) glowing neon katana centered on pure white background,
Panel 3) black armored chestplate centered on pure white background,
Panel 4) futuristic neon-lit Tokyo street at night,
Panel 5) cybernetic boots centered on pure white background,
Panel 6) robotic hawk centered on pure white background,
Panel 7) glowing holographic emblem centered on pure white background,
Panel 8) portrait of fierce cyber-samurai warrior centered on pure white background. Professional product photography style.
33
53
692
111,585
GPT 4o 🍔
Prompt in ALT:
![A flat branded paper folds itself into the full 3D shape of a [Big Mac], mid-motion. Paper colors match the ingredients and McDonalds brand primary color palette. Dramatic studio lighting, origami texture detail, gradient shadows, Japanese minimalism feel.](/pic/orig/media%2FGrZP0AvXQAEkk7H.jpg)
ALT A flat branded paper folds itself into the full 3D shape of a [Big Mac], mid-motion. Paper colors match the ingredients and McDonalds brand primary color palette. Dramatic studio lighting, origami texture detail, gradient shadows, Japanese minimalism feel.
25
41
684
63,600
Comparing AI Video Models: Creepy found footage
Pika 2.0
Kling AI 1.5
OpenAI Sora
Google Veo-2
Runway Gen-3
Hunyuan Video
Hailuo MiniMax
Luma Dream Machine
I ran the same prompt in each model 3-4x and chose the best results:
33
117
670
73,960
👉PROMPT: Photo of a cat's eye, intensely illuminated, capturing the intricate details of its iris and fur. The rest of the feline's face is hidden in shadows, creating a sense of intrigue and curiosity
41
72
637
38,091
Comparing the same prompt in 4 AI video models:
Kling AI 1.5
PixVerse
Hailuo
Gen-3
Prompt below:
44
88
669
78,559
🎇 Video Model Comparison: Text to Video
→ Veo 3
→ Kling 2.1 Master
→ Hailuo 02
→ Runway Gen-4
→ Seedance 1 Pro
→ Pika 2.2
I compared the same prompt on each model x 2-4 generations and chose my favorites below:
First-person perspective, 24mm lens simulation with motion blur and subtle head bob, bursting through a smashed window of a burning high-rise in hyperspeed as AI drones scan the area below; glass shards glint mid-air in bullet-time slow motion, sparks and rebar catching fire; wind howls past the camera with debris streaking across frame; gritty and kinetic, immersive audio and POV-style distortion, cinematic style.
If you try it on a different video model, share your result below 👇
47
91
668
65,114
She smiles as she turns her head slowly to look at something just off camera, then starts to walk down the hall behind her into a sun porch facing a beach at golden hour.
41
38
663
75,348
Tencent's Hunyuan Video is the latest open source AI text-to-video model, and the quality is mind-blowing.
5 Examples from my first results:
1. Prompt: Low-angle tracking shot following a sleek BMW motorcycles driving on a busy street in New York City. Street lights reflecting off wet pavement
37
99
658
80,687
GPT 4o Prompt: 3D Miniature Store
Check the ALTs for prompts & full prompt & tips below:
ALT Design a 3D miniature store shaped like a giant Starbucks® iced coffee cup, with a glowing neon “Sip & Chill” sign above the entrance. Incorporate tiny human figures enjoying the store, soft clay textures, and a pastel mint and cream color scheme. The scene should exude a whimsical and playful atmosphere, viewed isometrically with high detail, capturing the essence of the Starbucks® brand.
ALT Design a 3D miniature store shaped like a giant LEGO® brick, with a glowing neon “Build & Play” sign above the entrance. Incorporate tiny human figures enjoying the store, soft clay textures, and a bright primary color scheme. The scene should exude a whimsical and playful atmosphere, viewed isometrically with high detail, capturing the essence of the LEGO® brand.
ALT Design a 3D miniature store shaped like a classic Nintendo® controller, with a glowing neon “Power Up Zone” sign above the entrance. Incorporate tiny human figures enjoying the store, soft clay textures, and a red, gray, and white color scheme. The scene should exude a whimsical and playful atmosphere, viewed isometrically with high detail, capturing the essence of the Nintendo® brand.
30
71
656
59,489
I can't stop testing Higgsfield's camera motion mixes:
• 360 Orbit
• Bullet Time
Image: Mystic 2.5
21
66
653
34,190
Comparing AI text to video results:
• Gen-3
• Kling AI 1.5
• Luma Dream Machine
• Hailuo MiniMax
Same prompt for each model run 3 times, and I chose the best result for each:
"Wide shot of the towering glaciers of Patagonia. Condors soaring high above the high glaciers and penguins on ice floes. The birds appear smaller contrasting with the huge glaciers. Soft, diffuse light with occasional sunbeams breaking through moving clouds. Icy blues and whites with the dark silhouettes of the animals."
Which one do you choose?
35
102
625
55,842
Video Model Comparison: Img2Vid
• Hailuo 02
• Kling 2.1
• Veo 3
Sound generated with the videos on Kling & Veo 3.
Here's a quick comparison of 3 models using the same Midjourney image and prompt:
48
56
628
78,830
I just animated my Midjourney images with @runwayml Gen-3.
The new 'first or last frame' feature lets me tell a story that loops back to the beginning. Mind blown!
▫️The Awakening▫️
33
101
619
56,796
Out of the shadows.
ALT A person's shadow detaches and glitches, leaving trails of pixelated artifacts.
18
84
586
41,683
Comparing AI video results:
• Gen-3
• Kling AI 1.5
• Hailuo MiniMax
• Luma Dream Machine
I used a FLUX1.1 [pro] image as the input. Each model ran the same prompt 3 times, and I chose the best results:
"Smooth side tracking shot of a woman jogging at twilight with the city skyline in the background, sharp focus on the woman as she pumps her arms with each stride as she jogs, depth of field, advertising commercial"
Running, jogging, and walking are difficult for AI video models, and none of them produced a smooth result, but this is the worst they will be.
38
83
598
57,660
Nano Banana Image to Video 3-Step Prompt Worklfow 🍌
Step 1 - Grid theme prompt:
8-panel grid layout:
Panel 1) flowing black velvet cloak on white background,
Panel 2) silver gothic pendant centered on white background,
Panel 3) dark leather boots centered on white background,
Panel 4) candlelit gothic castle hall with stained glass windows,
Panel 5) goblet of blood-red wine on white background,
Panel 6) black raven centered on white background,
Panel 7) ornate dagger centered on white background,
Panel 8) mysterious pale figure portrait on white background. Each item distinct, styled as product photography.
22
75
614
70,563
🚨 How does Adobe Firefly Video stack up against other AI video models?
I ran the same prompts on:
• Runway Gen-3
• Hailuo AI’s MiniMax
• Luma AI Dream Machine
The results were fascinating.
5 Examples below:
34
95
595
73,152
How to train a FLUX LoRA on an illustrated character sheet to generate realistic, consistent characters.
Workflow below:
24
81
590
55,013
🚨 Character Consistency in @Kling_ai Video
I got early access to "Custom Models" and the results are mind-blowing.
Train a model on your face to generate videos in any setting.
Workflow below:
76
59
572
77,194
I still can't use Sora.
But I generated this video from my Midjourney image with Stable Video for free.
Plus 7 more impressive examples from the website:
22
51
546
62,420
This is very cool!
Lip sync in Veo 3 image to video
Prompt: She sighs and says "It's about time! You can finally lip sync characters in Veo 3!"
36
53
576
55,405
I got early access to @hedra_labs Character-1 and it's incredible.
The model generates video and dynamic 3D content with a focus on expressive humans.
Workflow and examples below:
#hedra
47
86
553
108,273
Not gonna lie - @runwayml Act-One is impressive:
51
52
554
54,823
Sora who?
Generate your own videos from text or your favorite images right now with @LumaLabsAI Dream Machine!
10 of my favorite examples with prompts:
#LumaDreamMachine
38
63
553
99,578
🔹PAPER QUILLING KIRIGAMI🔹
I tested some of my old Paper Art prompts in Midjourney V6.
Here's some of my amazing results with prompts so you can try them too:
42
59
534
38,118
👉Prompt: A koi fish, its scales a mosaic of orange and white dots, appearing to swim across a canvas of deep blue dotted waves, infinite reflection
49
64
509
26,134
GM 🌞
Finally, Friday!
Drop your favorite AI art you created this week 👇
ALT A woman in an afro - caribbean style with a big headpiece, in the style of neon art nouveau, dark cyan and light amber, flowerpunk, layered images, monochromatic shadows, light orange and black, romantic illustration
137
54
481
63,349
This is wild.
I trained a Flux.1 Dev model with my headshots on @Astria_AI
Guess which one is the real me - three of these are AI-generated 👇
134
23
539
77,806
I generated an AI music video with Plazmapunk Beta.
Upload your own music or choose from the selections. You don't upload your images, so you pick from the options available.
Enter a short text description of your video & generate.
Link in the next tweet 👇
38
81
524
196,545
🎆 Video Model Comparison: Text to video
There are a few new models in this comparison, including Wan 2.5, Kling 2.5 & Luma Ray3:
→ Wan 2.5
→ Kling 2.5
→ Luma Ray3
→ Veo 3
→ Seedance Pro
→ Hailuo-02
→ Runway Gen-4
→ PixVerse 5
I used the same prompt on each model up to 4 times:
First-person POV racing on a motorbike through a mountain road, leaning into sharp turns, wind rushing past, headlights streaking across the dark tunnel ahead.
Upscaled with @topazlabs Astra:
33
88
563
45,232
I've been experimenting with this prompt, including Zooming and Panning in Midjourney and it produces beautiful results.
Here are 7 of my favorites:
(Prompt in thread)
43
77
511
111,158
▫️LINE ART▫️
PROMPT: a minimalist line drawing of a mountain range with a rising sun, the parallel lines in white with a hint of sunrise pink on the highest peak
33
52
538
56,302
I discovered a secret to unlocking unlimited variations in Midjourney.
I used Vary Region inpainting differently than most people.
I'll show you a peak at my new workflow:
33
49
499
144,670
Video Model Comparison: Image2Video
Same input image + text prompt on each model:
• Pika 2.0
• OpenAI Sora
• Runway Gen-3
• Kling AI 1.6
• Luma Dream Machine
• Hailuo MiniMax
I used the same prompt that I used to generate this image in Midjourney and chose the best results for each model tested.
The prompt didn't include specific camera motion, so it was interesting to see each model's interpretation:
46
73
519
63,105
This is not real.
Completely text to video generated by @LumaLabsAI Ray2.
I used ~ 30 video clips about Marine Biology, sound effects by Epidemic Sound, edited with Davinci Resolve:
55
70
497
38,639
Vintage.
ALT A vintage train steaming down the tracks in grayscale, the billowing smoke turning into an explosion of watercolor blues and grays, the colors dripping down into a reflective puddle along the railway --sref 1975031384
22
55
452
12,722
Reve Image 1.0 is really good.
So I compared it to 5 other image models:
• Midjourney
• Hailuo AI
• FLUX 1.1
• Mystic 2.5
• Imagen 3
Examples & link👇
37
72
516
66,577
🎥 Comparing AI video models: Image to video
• Gen-3
• Kling AI 1.5
• Hailuo MiniMax
• Luma Dream Machine
I used a Midjourney image in each model 4 times with no text prompt.
This type of image is difficult for the AI to separate the subject in the front from the people behind her - they tend to move the group as if they are a single unit.
But the results were interesting, as you can see!
I chose my favorite results, below.
43
54
504
41,356
Imagine ChatGPT, Midjourney, and Powerpoint merged into one.
That's Gamma's latest update.
If you're craving an all-in-one platform, you need to see this:
42
75
492
126,516
🔹Comparing AI video results:
• Gen-3
• Kling AI 1.5
• Hailuo MiniMax
• Luma Dream Machine
I used a FLUX LoRA trained on my selfies as the input. Each model ran the same prompt 3 times, and I chose the best results:
"Tracking shot of a curator standing and speaking to the camera as she welcomes museum guests. Warm, inviting atmosphere."
Each model interpreted the prompt differently, and the backgrounds are interesting:
52
67
497
41,295
// LINE ART //
Prompt: Illustrate an elegant cat in profile using sleek white lines on a black canvas, with just a few lines in sapphire blue highlighting its eyes
72
46
472
35,881
🔳QT Your Black & White🔳
Prompt in ALT:
ALT Model in a minimalist outfit consisting of a sleek black turtleneck and high-waisted trousers, positioned against a white background to contrast the simplicity and elegance of the attire.
19
70
445
22,327
I built a cinematic video prompt generator with @AnthropicAI Claude Artifacts.
I tested the prompts in @runwayml Gen-3 and the results are impressive.
Here's an example followed by a demo of the prompt generator:
58
63
503
61,077
💠 QT Your Koi
Prompt in ALT:
ALT A koi fish, its scales a mosaic of orange and white dots, appearing to swim across a canvas of deep blue dotted waves, infinite reflection, prismatic
15
38
442
27,422
Image editing just evolved.
Flux.1 Kontext turns LTX Studio into a multimodal, context-aware platform.
A leap forward for filmmakers, advertisers & pro creatives ↓
18
51
490
32,604
Kling is cooking and it's looking good.
I got early access to the newest model to be released soon:
56
34
487
23,227
Midjourney -> Kling 1.6
Prompting for nature scenes seems complicated but you really only need to describe what you want to see in your video.
If you have an input image, think about what would actually occur in real life and include the details in your prompt:
• Subject and surrounding areas
• Time of day, lighting, and weather
• Environmental interaction with objects from wind, rain, snow, etc.
Finally, describe the type of camera motion to complete the scene.
I used this prompt to animate my Midjourney image with Kling 1.6:
Cinematic wide shot of a beautiful contemporary lakefront cabin in a redwood forest on a sunny day. Camera pushes in very slowly as the sunlight filters through the tree canopy and reflects off the lake water. A light breeze blows the trees and water.
Share your nature scenes below👇
31
53
486
31,601
Another fun and easy prompt keyword to produce beautiful Midjourney images:
"Luminogram"
I used the same prompt for 7 colorful images in this thread. I'll show you how I did it:
(Prompt is in the thread)
1.
33
43
467
113,363
It's my birthday, so I'm going through some of my favorite images and videos over the last year🥳
I was thrilled to be able to add myself into scenes with Midjourney's Character Reference several months ago:
198
22
464
17,335
Do you love trying new AI art styles but don't know where to find them?
You're wasting time by looking in the wrong places.
This is where I go first:
@Midlibrary_io
40
77
449
73,301
Tried multi-shot prompts with just one image using Seedance 1.0
This format from @framer_x works great for storytelling. Each [cut] shifts the shot.
Here's a video made from three scenes I wrote around one little character's walk:
28
46
477
35,878
Very impressive 🔥
RMBG 2.0: SoTA open source background removal model by @bria_ai_
I chose an image with similar colors on the subject and background - it removed the background perfectly.
Link below.
17
76
473
39,820
Don't be afraid to step outside of your comfort zone.
You never know what you might achieve.
Midjourney -> Kling 1.6
34
31
450
37,257
Finally, my first result from @Kling_ai!
I animated my Midjourney image:
64
47
431
32,615
I got a call in the middle of the night from my son - his car spun out of control in heavy rain.
He's a safe driver and was completely sober. Thank God he's alive with only minor injuries.
Kiss your loved ones and remember tomorrow isn't promised to anyone.
Live your life.
152
6
435
13,756
I just fine-tuned a Flux.1 Dev LoRA with my own images on @Astria_AI.
The results are incredible!
Workflow and tutorial👇
40
57
440
62,751