The Agentic Creative Studio that remembers how you create. Brief · Run · Edit on canvas Join us Discord: discord.gg/44qN9MtDHb
- Tweets 272
- Following 51
- Followers 467
- Likes 406
Pinned Tweet
Hy Image3.5 preview is live on Miora. FREE through October 7.
Campaign visuals. Product scenes. Storyboards. Game concepts.
Up to 2K. Text-to-image and image-to-image.
Try it: miora.design/
13
9
92
57,374
Upload a product image and place it in a new setting with Hy Image3.5 preview. Keep key details intact with more consistent edits.
Try it free through Oct 7: miora.design/
49
Less “hope the prompt works.”
More “make it exactly like this.”
Nice work from @HeyAmit_ exploring the control you get with Hy Image 3.5.
Try it for free in Miora through October 7.
I tested Tencent's Hy Image3.5 preview.
What stood out wasn't just the image quality.
It's how much control you get over things like text, references, realism, and editing.
Here's what I found 👇
1
108
If you’re going to try it, you might as well throw the whole creative brief at it. @altiamkabir understood the assignment.
Go make something fun with it in Miora, free through Oct 7.
🚨🚨 FREE AI IMAGE GENERATION JUST GOT SERIOUS.
HY Image3.5 Preview is FREE to use right now. Create posters, e-commerce visuals, UI designs & photorealistic ads.
I tested it inside @Miora_Design. Here’s what I created 🧵👇
1
4
195
Fast is good.
Fast and powerful is where things get fun.
Jump in and make something, free through Oct 7.
While Qwen and DeepSeek are teasing 2T-8T models, @TencentHunyuan is heading to a counterintuitive direction - be small but powerful. I’ve been using HY4 Preview and it’s insanely fast and good. This worths a read 👇👇
145
Add up to 5 reference images and give your game world a clear direction. Build character, environment, and UI concepts with Hy Image3.5 preview.
Try it free through Oct 7: miora.design/
1
3
143
Miora Design retweeted
🎉 More ways to create with Hy Image3.5 preview
✨ Miora | Create with more accurate text rendering and more realistic visual detail【Free till Oct 7】
⚡ WAND | Generate images directly in VOD and MPS with one-click access and streaming output.
🤖 Tencent Cloud ADP | Build image-generation agents with natural language, then publish via APIs or shareable links.
🔑 TokenHub | Access Hy Image3.5 preview for your applications and workflows.
See Hy Image3.5 preview in action ⬇️
1
1
338
Kuria came to stress-test, not play nice.
Product ads, posters, typography, game UI, environments, edits… Hy Image 3.5 got the full workout.
Give it a spin in Miora, free through October 7.
I gave Tencent’s new Hy Image3.5 preview some serious image-generation tests.
Not just portraits.
I threw at it:
• Product ads
• Commercial posters
• Complex typography
• Game UI
• Detailed environments
• Reference-image editing
Here’s what happened.
1
2
280
Big ideas, lots of branches, and some seriously ambitious world-building. Huge congrats to the team behind “The Only One Awake.”
Feeling inspired? Jump into Miora and start building. Free through Oct 7.
Animation Champion! 15-Year-Old Hong Kong Students Built an Interactive Cinematic Game with AI
From one animation to an interactive cinematic game with 37 scenes, 10 choices, and 7 endings — the Animation Track Champion of the Tencent Cloud Hackathon Global Finals is The Only One Awake!
Three 15-year-old students from Hong Kong used Miora for cross-scene style anchoring and asset consistency, and WorkBuddy for visual narrative mapping, crafting a dystopian allegory interactive cinematic game where every choice leads to a different ending.
With AI, they made their story not just watchable — but playable, choosable, experienceable.
#TencentCloudHackathon #Hackathon #AnimationTrack #Miora #WorkBuddy #InteractiveCinematicGame #HongKongStudents #AIAnimation #TechforGood
2
191
Good visuals?
Check.
Precise text?
Also check.
Nice work from @Alex_Inspira taking Hy Image 3.5 through a full campaign workflow.
Your turn. Try Miora free through Oct 7.
LOS MODELOS DE IMAGEN CON IA SUELEN OBLIGARTE A ELEGIR:
BUENOS ELEMENTOS VISUALES O TEXTO PRECISO.
Hy Image3.5 preview consiguió ambas cosas en un solo flujo de trabajo.
Lo usé para crear y editar con precisión una campaña completa para una bebida energética dentro de Miora.
Aquí tienes el proceso completo ↓
147
Turn your next character, storyboard, or promo art idea into an image. Explore richer realism and more visual styles with Hy Image3.5 preview.
Try it free through Oct 7: miora.design/
2
4
394
This is the kind of test that shows whether a model has range, not just a signature look.
Tencent just released Hy Image3.5 preview, their new image generation model available inside Miora.
Compared to previous iterations, the main improvements here focus on physical realism, complex prompt following, and dense typography across both English and Chinese.
I tested it across a few completely different styles to see how it holds up. Starting with macro wildlife photography: every individual scale and water droplet renders with authentic depth of field instead of typical AI plastic smoothing.
1/7
1
3
314
Posters. Game worlds. Brand identities.
See where your visual language can go with Tencent Hy-Image 3.5 Preview on Miora.
Start free through Oct 7 → miora.design
#Miora #MioraDesign #HyImage35 #AIDesign #GenerativeAI
2
10
886
One good image is easy. Keeping a product consistent across an entire campaign is the real challenge.
Nice work from @KaiteeShiks, pushing Hy Image 3.5 beyond the one-shot demo.
Most image models can make one beautiful ad.
I wanted to see whether Hy Image3.5 preview could handle the harder part: carry one product through an entire campaign without losing its packaging, typography or visual identity.
So I built NORTHLINE from scratch.
@TencentHunyuan @TencentAI_News @Miora_Design
5
406
Generation is easy. Knowing what’s worth making is the harder part.
A great perspective from Luca Maxim, creator of Children of Khan, on why human insight & creativity still has to lead the process.
2
3
519
A proper test drive: text-heavy posters, product edits, and UI.
Shoutout to @AIwithShoaib for putting Hy Image 3.5 through a hands-on workout.
Tencent built an image model that actually spells its own text right.
Hy Image3.5 preview is live in @Miora_Design today. I got early access before the crowd.
Three things I threw at it:
→ a poster packed with text
→ a product image edit
→ a full UI screen
Here’s what came back 🧵
1
2
6
1,290
Our 3rd place winner from the Bestiary: Miora × MiniMax H3 Creative Contest
Faith withers when death knocks at the gate.
《焚香剖佛》 (The Desecrated Lotus) A traditional Chinese 2D animated dark fable exploring greed, sacrilege, and karmic transmutation.
Crafted for the Bestiary Creative Contest with Miora & MiniMax H3.
@Miora_Design @Hailuo_AI
#MioraDesign #MioraBuddy #TencentAI #MiniMaxH3 #MiniMax
1
4
179
Miora Design retweeted
A professional image model that doesn't cost like one. $0.024 per 2K image.
Meet Hy Image3.5 preview.
> up to 5 reference images in, and the meter only runs on what comes out
>multi-turn editing with earlier edits in context
> text, layout and 2K consistency, the parts that break posters and UI mocks
Free for two weeks in @OnSoloAI and @Miora_Design .
API:console.tencentcloud.com/tok…
Previews exist to get corrected, tell us what's off and it goes into the next version.
9
8
92
29,452