Katalina Ortuno

🏋🏽‍♀️🎾Weightlifting & Tennis lover / 💻Senior GTM 💜

[Katalina Ortuno @92katortuno]

“In a world obsessed with being extraordinary, there is a deep peace in just being a good human”
someone I wish I knew

[Pruna AI @PrunaAI]

FREE VIDEO GENERATION, to celebrate our launch!

• Until Friday 9pm CET, p-video is free for all within rate limits.
• Until Monday 9pm CET, post one nice p-video creation and DM us to get 10$ in free credits on our API.

Generate with P-video and share your creations!

[Replicate @replicate]

P-video by @PrunaAI is here.

10¢ for 10 second video, done in < 10 seconds, with native audio.
The video below was made using P-video.

And it's free for the first 24 hours!

[Pruna AI @PrunaAI]

Happy to announce P-Video, delivering the fastest and cheapest video generation on the market.

• 𝗛𝗶𝗴𝗵 𝘀𝗽𝗲𝗲𝗱: 10s generation speed for 5s of 720p, 18s generation speed for 5s of 1080p video.
• 𝗔𝗳𝗳𝗼𝗿𝗱𝗮𝗯𝗹𝗲: $0.02 per second of 720p video, $0.04 per second of 1080p video.
• 𝗛𝗶𝗴𝗵 𝗰𝗼𝗻𝘁𝗿𝗼𝗹: A unique draft mode for video preview, taking 2.5s to preview 5s of 720p video and 4.9s to preview 5s of 1080p video.
• 𝗠𝘂𝗹𝘁𝗶-𝗺𝗼𝗱𝗮𝗹: Support text, image, and audio input for up to 15s video generation for any format, with audio output support.

💨 Run it self-serve via API here: https://t.co/DVBQ8K71iX - https://t.co/DVBQ8K71iX -, also available on @Replicate, @Runware, @Cloudflare, @runpod, @WiroAI, @Qubrid_AI
👉 Test it in the playground: https://t.co/AfnYID0kI2 - https://t.co/AfnYID0kI2 -
🎡 Explore applied demos: https://t.co/8Q2f4UfH3U - https://t.co/8Q2f4UfH3U -
⭐️ Star our OSS GitHub: https://t.co/8OGBDxHpzo

[Arena.ai @arena]

📉🖼️Image Arena Pareto Frontier

Image use cases vary widely. Sometimes you want the highest quality, and sometimes you need something efficient enough to run at scale.

Looking at Arena Score versus price per image lets us see which models sit on the Pareto frontier.

Top models on the Pareto frontier for Text-to-Image:
- @OpenAI: GPT‑Image‑1.5‑High‑Fidelity, GPT‑Image‑1‑Mini
- @bfl_ml: Flux‑2‑Max, Flux‑2‑Flex, Flux‑2‑Pro, Flux‑2‑Dev
- @GoogleDeepMind: Nano‑Banana
- @TencentGlobal: Hunyuan‑Image‑3.0
- @PrunaAI: P‑Image

[Pruna AI @PrunaAI]

Imagining change? Believe this little boy! We just dropped a Wan 2.2 Animate Replace update, and it is the fastest and cheapest option out there!

- It is now 4x faster, so it only takes a handful of seconds
- cleaner pricing! 0.02$/s for 480p and 0.05$/s for 720p

Now available on https://t.co/7kp1HKS820: https://t.co/rF1T5KYxXN

[Katalina Ortuno @92katortuno]

RT @PrunaAI: 🚀 𝗧𝗵𝗲 𝗳𝗮𝘀𝘁𝗲𝘀𝘁 𝗸𝗻𝗼𝘄𝗻 𝗚𝗣𝗧-𝗢𝗦𝗦-𝟭𝟮𝟬𝗕 𝗲𝗻𝗱𝗽𝗼𝗶𝗻𝘁 𝗼𝗻 𝗮 𝘀𝗶𝗻𝗴𝗹𝗲 𝗕𝟮𝟬𝟬 𝗶𝘀 𝗻𝗼𝘄 𝗹𝗶𝘃𝗲 — 𝗱𝗲𝗹𝗶𝘃𝗲𝗿𝗶𝗻𝗴 ~𝟮× 𝗯𝗲𝘁𝘁𝗲𝗿 𝘀𝗽𝗲𝗲𝗱 𝗮𝗻𝗱 𝗿𝗲𝘀𝗽𝗼𝗻𝘀𝗶𝘃𝗲𝗻𝗲𝘀𝘀 𝘁𝗵𝗮𝗻 𝗰…

[Pruna AI @PrunaAI]

To celebrate together the release of FLUX.2 [flex] in collaboration with @bfl_ml. We’re launching the FLUX.2 [flex] Design Challenge! 🎨

𝗧𝗵𝗲𝗺𝗲
𝗠𝗮𝗸𝗲 𝗮𝗻 𝗶𝗻𝗳𝗼𝗴𝗿𝗮𝗽𝗵𝗶𝗰 𝗼𝗳 𝗵𝗼𝘄 𝘁𝗼 𝗴𝗿𝗼𝘄 𝗮 𝗯𝗹𝗮𝗰𝗸 𝗳𝗼𝗿𝗲𝘀𝘁 𝗼𝗳 𝗽𝗹𝘂𝗺𝘀

𝗛𝗼𝘄 𝘁𝗼 𝗽𝗮𝗿𝘁𝗶𝗰𝗶𝗽𝗮𝘁𝗲
1️⃣ Answer with a single creative infographic using FLUX.2 [flex] and the screenshot showcasing it. You can try it here: https://t.co/Aq3z1zhWk0
2️⃣ Follow the @PrunaAI and @bfl_ml account in X.
3️⃣ Mention us and add the hashtag #flux2flex.

𝗣𝗿𝗶𝘇𝗲
Popularity + Judge’s evaluation
🥇 €150 🥈€100 🥉 €50

𝗗𝗮𝘁𝗲𝘀
Until February 6th (23:59 CET)

𝗝𝗼𝗶𝗻 𝘁𝗵𝗲 𝗰𝗵𝗮𝗹𝗹𝗲𝗻𝗴𝗲, 𝘃𝗼𝘁𝗲 𝗳𝗼𝗿 𝘆𝗼𝘂𝗿 𝗳𝗮𝘃𝗼𝗿𝗶𝘁𝗲𝘀, 𝗮𝗻𝗱 𝗶𝗻𝘀𝗽𝗶𝗿𝗲 𝘁𝗵𝗲 𝗰𝗼𝗺𝗺𝘂𝗻𝗶𝘁𝘆!
Read the rules in detail: https://t.co/xLbWwWgnDe

[Pruna AI @PrunaAI]

FLUX.2 [flex] is now 3x to 4x faster ⚡

Working with @bfl_ml, we accelerated one of their top-ranked image model to ~5s per image generation and ~10s per image edit, making it the de facto standard for typography-heavy image generation.

The results:
- On one H200: ~4x faster
- On four H200s: ~3x faster
- Image generation in 13s/5s on single/multi-gpu
- Image edits from in 28s/10s on single/multi-gpu
- Leads the Pareto front for speed vs quality

Perfect for designers and teams who need fast iteration on typography-heavy visuals without compromising quality.

This is what AI efficiency looks like when done right: efficiency without quality loss!

Read our blog: https://t.co/XgfyspliZR
Try the official accelerated endpoint: https://t.co/Aq3z1zhWk0

[Black Forest Labs @bfl_ai]

Our best model for text and typography, FLUX.2 [flex], is now up to 3x faster.

FLUX.2 [flex] delivers highly accurate text rendering, custom typography, richer colors, and smoother visuals. We worked with @PrunaAI to make it up to 3x faster.

Examples in the thread:

[Katalina Ortuno @92katortuno]

If someone tells you they have AI GTM fully figured out, call them on their BS.

Certainty is expensive in AI.

I wrote about dropping GTM theater and what working at my first AI company is forcing me to unlearn.

https://t.co/r3krcyZnW7

[Pruna AI @PrunaAI]

Flux 2 Klein from @bfl_ml is super small — and now super fast on @replicate⚡️

- Base Flux 2 Klein: Generate a 1MP image in 1.3s, edit a 1MP image in 2.7s!
- Distilled Flux 2 Klein: Generate a 1MP image in 0.3s, edit a 1MP image in 0.7s!
- Also supports multiple image references ;)

Awesome collaborating with @bfl_ml to optimize their models — more coming soon 👀

👉 Try them here:
- Base 4B: https://t.co/xdphMik22N
- Distilled 4B: https://t.co/1aewN3WAkC
- Base 9B: https://t.co/6Q2LrZ4M53

🔓 Open-source support coming soon in pruna: https://t.co/8OGBDxHpzo

[Pruna AI @PrunaAI]

🎄 𝗪𝗲 𝗯𝘂𝗶𝗹𝘁 𝟮𝟰 𝗚𝗲𝗻𝗔𝗜 𝗰𝗼𝗻𝘁𝗲𝗻𝘁 𝗮𝗽𝗽𝘀 𝗶𝗻 𝟮𝟰 𝗱𝗮𝘆𝘀. 𝗛𝗲𝗿𝗲'𝘀 𝘄𝗵𝗮𝘁 𝘄𝗲 𝗹𝗲𝗮𝗿𝗻𝗲𝗱!

The driving force was our performance models, P-Image and P-Image-Edits:
• Real-time image generation & editing with sub-second inference
• Negligible costs for only $0.005 per generation and $0.01 per edit.

The lessons we learned mostly stem from our performance models, but there are some related to prompting, too:
• Speed matters more than perfection, with a quick and cheap generation and edits, you can make (almost) anything work.
• Proposed prompts from Claude and the like work as a head start but don’t overcomplicate them!
• Minor tweaks in phrasing sometimes make all the difference, e.g., “replace” vs “swap” vs “change” all sound similar but might have completely different downstream effects.
• Beware of the sunk cost fallacy and don’t feel bad to restart fresh and simple.

We’ve obviously learned more, but stay tuned to see how this will be integrated into what we build and ship next.

Try all demos: https://t.co/NfSXSciTUJ

[Pruna AI @PrunaAI]

🎉 We made the new 𝗤𝘄𝗲𝗻-𝗜𝗺𝗮𝗴𝗲-𝟮𝟱𝟭𝟮 ultimate fast on @replicate for the new year!

The @𝗤𝘄𝗲𝗻 𝘅 @𝗣𝗿𝘂𝗻𝗮 collaboration continues after 𝗤𝘄𝗲𝗻-𝗟𝗮𝘆𝗲𝗿𝗲𝗱 and 𝗤𝘄𝗲𝗻 𝗜𝗺𝗮𝗴𝗲 𝗘𝗱𝗶𝘁 𝟮𝟱𝟭𝟭 ⚡️

✅ Generates high resolution images in ~𝟳𝘀
✨ Realistic humans without “AI look”
🧑‍🎨 Fine textures for any materials like glass, water, fur
⌨️ Superior text rendering with adjusted layouts and accurate composition

Try it 👉 https://t.co/7iVIQpEo3a

[Pruna AI @PrunaAI]

Happy to announce 𝗣-𝗜𝗺𝗮𝗴𝗲 and 𝗣-𝗜𝗺𝗮𝗴𝗲-𝗘𝗱𝗶𝘁, built for performance. They enable real-time image generation and editing.

You can now:
• Generate an image for $𝟬.𝟬𝟬𝟱 in 𝟭 𝘀𝗲𝗰𝗼𝗻𝗱
• Edit an image for $𝟬.𝟬𝟭 in 𝟭 𝘀𝗲𝗰𝗼𝗻𝗱
all without compromising quality. The models deliver strong prompt adherence, precise text rendering, fine-grained details, multi-reference image capabilities, and more.

Test it in the playground: https://t.co/h5cpGgCjwE - https://t.co/h5cpGgCjwE -
Explore applied demos: https://t.co/8Q2f4UfH3U - https://t.co/8Q2f4UfH3U -
Model pages:
• https://t.co/A2CeHwjfFV - https://t.co/A2CeHwjfFV -
• https://t.co/ajE8F4OVW9

[Jim Laurie @laurie_jim]

🎉 Happy Birthday Strapi Cloud ☁️
Today is the first anniversary of Strapi Cloud!

It has been so far a super success, really happy about what we've done here 🚀
Deploy your Strapi application in a few clicks, we manage everything 💪

Give it a try 👇
https://t.co/oGXqN9Z0kb

[Maeva Lienard @MaevaLienard]

I’m thrilled to share an article that I wrote about a topic dear to my heart: Open-Source. For this one, I decided to deep dive into the subtleties of designing Open-Source products ✨

https://t.co/fhi0E18wa5

[Camille Charrière @camtyox]

POV: ravioli

[Jim Laurie @laurie_jim]

🔔 October @strapijs Community call is tomorrow 🗓️

Last chance to register 👇
https://t.co/pb0NNIRQDO

See you tomorrow 👀

[Strapi @strapijs]

Mark the date: On October 10th, we're going live to unveil our 2024 Roadmap, explore the latest updates including Strapi V5, highlight community initiatives, and answer all your questions!

👉 Join us for our upcoming Community Call:
https://t.co/3AqUlDnZFN