Sara Han

DevRel & ML | 🐕🎮💪💻

[Johanna Sommer @j_m_sommer]

If you don't want to go through the traumatic experience of trying on clothes during a 35-degree European summer, check out p-image-try-on 👀 It's really good and also wayyy faster than changing irl

[Wiro @WiroAI]

📣 P-Image-Try-On is now available on Wiro AI.

The fastest and most affordable virtual try-on model on the market.

— $0.015 for the first garment, $0.008 for each additional garment
— Try on up to 11 garments in a single generation
— Quality mode in under 2 seconds per garment
— Turbo mode in under 4 seconds total
— Prompt and pose guidance for precise control

Built for fashion, e-commerce, marketplaces, and catalog teams shipping virtual try-on experiences without costly reshoots.

Available now on Wiro AI.

[Replicate @replicate]

P-image-try-on from @PrunaAI is now on Replicate!

The most affordable virtual try on model out there.

[Sara Han @sdiazlor]

Forget about the traditional white background photo with garments to select your outfit; it's time to move on.

P-Image-Try-On by @PrunaAI!
Select your background, pose, and garments, and see how it actually looks on you in just 2s!

Try it now: https://t.co/jmX9XJQCTy

[Pruna AI @PrunaAI]

P-Image-Try-On delivers the fastest and cheapest virtual try-on on the market.

• Affordable: $0.015 for the first garment, $0.008 for each additional one
• Multi-garment: fit up to 11 garment references onto a person photo
• Quality mode for <2s per garment
• Turbo mode for even higher speeds <4s in total
• Prompt guidance and pose guidance for precise control

Ideal for fashion, e-commerce, and catalog teams shipping VTO without reshoots.

Available day 0 on Pruna API and with our launch partners:

@cloudflare @scenario_gg @replicate @wiroai @tellersai @eachlabs @wavespeed_ai @runware @LookFashionai

👉 Playground: https://t.co/fgi0HJ02oY
📚 API: https://t.co/ahqlgpbJoW
🧩 P-Video, P-Image, P-Image-Edit & more: https://t.co/kFmfarZaA1

[Sebastian Raschka @rasbt]

It's been a while! 4 nice additions to the open-weight local-LLM-on-consumer-hardware ecosystem:

[Pruna AI @PrunaAI]

P-Video-Replace is now the fastest model for top-quality character replacement in existing video.

- $0.03/s at 720p and $0.06/s at 1080p — a small fraction of typical character-replacement / video-edit pricing (~$0.07–$0.35/s at 720p in our benchmarks)
- 3.58 s generation time per second of output video at 720p — roughly ~10× faster than typical alternatives in the same benchmarks (~36–43 s per second of output)
- Upload a source video, 1–4 reference character images, and an instruction prompt to swap the character in the source video while keeping the rest unchanged.

Ideal for UGC ad variations, meme remixes, film recasting, gaming cinematics, and educational videos when you already have a winning take and need a new face, creator, or persona in the same shot.

Try it in the playground: https://t.co/My2rZVFjXm

We're launching with 70% discounted usage which ends on 2026-06-08 00:00 CEST!

Try it at any of our launch partners: @PrunaAI, @eachlabs, @inference_sh, @replicate, @runware, @scenario_gg, @tellersai, @togethercompute, @wavespeed_ai, @wiroai

Also available on the Pruna API: https://t.co/3wdrtRPtil

We'd love your feedback as you try it.

📚 Model docs & API examples: https://t.co/ovGHToyvjx
📚 API quickstart: https://t.co/LXjHAjqkKC
⭐️ OSS: https://t.co/hCo1AtzIjF
🌐 https://t.co/bLLLZPE0FI

[Sara Han @sdiazlor]

Last week, P-Video-Animate let you bring a photo to life. But what if you wanted to keep the original video’s aesthetic while replacing only the character?

Today, @PrunaAI releases P-Video-Replace ✨️

Learn more: https://t.co/iQctduk6Gm

[Sara Han @sdiazlor]

@PrunaAI https://t.co/yWG3TCZ5VB

[Sara Han @sdiazlor]

𝗣-𝗩𝗶𝗱𝗲𝗼-𝗔𝗻𝗶𝗺𝗮𝘁𝗲 by @PrunaAI is released!

Source video + reference image = an animated image with the source video’s motion, timing, movement, and scene.

And with the best speed: 𝟱.𝟮𝟰𝘀/𝘀

Try it through the Pruna API:

[Pruna AI @PrunaAI]

P-Video-Animate is now the fastest model for top-quality image-to-video motion transfer.

- $0.03/s at 720p and $0.06/s at 1080p — a fraction of typical motion-transfer pricing (~$0.07–$0.35/s at 720p in our benchmarks)
- 5.24 s generation time per second of output video at 720p — roughly 7× faster than typical motion-transfer alternatives in the same benchmarks
- Upload one still plus a driver video to preserve motion, timing, camera movement, and scene structure, with optional source audio and clips up to ~2 minutes at 720p
Ideal for UGC ads, gaming cinematics, film casting, and product demos when you already have a winning take and want a new hero still to follow it.

Try it in the playground: https://t.co/kUV7WzUXyd

We’re launching with 70% discounted usage for the coming two days with our launch partners.

@eachlabs, https://t.co/YVg2KQrDmD
@replicate, https://t.co/By5Qk4ovrI
@runware, https://t.co/gqhDs5yASR
@scenario_gg, https://t.co/UQI8NF8jyR
@wavespeed_ai, https://t.co/HXrPVNHtRf
@wiroai, https://t.co/fmxGinBBiu
@TellersAI, https://t.co/DiBR9LIP16

Soon also on @togethercompute

We’d love your feedback as you try it.

📚 Model docs & API examples: https://t.co/Tv5x6TXqlR
📚 API quickstart: https://t.co/fwgnfWSXIh
⭐️ OSS: https://t.co/8OGBDxHpzo
🌐 https://t.co/DU9Bj08Q5R

[Pruna AI @PrunaAI]

Pruna OSS Community Spotlight: Token Merging by rensortino

Token Merging is now available for vision transformer models. This algorithm progressively merges similar tokens between the attention and MLP stages of each transformer block, reducing the number of tokens processed during inference. The result: faster inference with minimal impact on quality.

Thank you for contributing to Pruna OSS!

⭐️ Star our OSS: https://t.co/9yGJ3fRq2w
📚 API & quickstart: https://t.co/ybGIDGss1Q
🌐 Website: https://t.co/DU9Bj08Q5R

[Pruna AI @PrunaAI]

🌊 We know it took a while, but we we are proud we made it happen. As of today, all of our models are available on @wavespeed_ai!

Find the models on their website: https://t.co/clPxW5J0hu

As fast and as cheap as always!

[WaveSpeedAI @wavespeed_ai]

Excited to announce our official collaboration with @PrunaAI 🎉
Their powerful image & video models are now live on WaveSpeedAI

✨ High-quality visual generation
🎬 Cinematic motion & consistent characters
🖼️ Text-to-Image, Text-to-Video & Image-to-Video
⚡ Fast generation built for scalable creative workflows

Go try the Pruna AI models on WaveSpeedAI 🚀

[Pruna AI @PrunaAI]

P-Video-Avatar is the fastest and most cost-effective avatar video model available, and launching end-to-end generation workflows has never been easier!

Quality on par with Veo 3.0, 18x faster and 6x cheaper than alternatives like Fabric, OmniHuman, and HeyGen.

Learn more about the model: https://t.co/t3Q4l4oP2l
Check out the @Inferencesh skill: https://t.co/zJ6Xaqypoj

[John @johnrachwan]

I'm so excited to announce P-video-avatar.

It is the first model we've released that is not only the best in terms of speed and cost but also at the frontier in terms of quality for the avatar usecase.

[Sara Han @sdiazlor]

P-Video-Avatar has just launched, ready to turn a single portrait into a talking-head video. Use your own script or upload audio, choose from 30 voice options across multiple languages, and control the tone, pacing, and emotion. And the best part? It will be really fast👇

[Pruna AI @PrunaAI]

P-Video-Avatar is now the fastest model for top-quality and diverse avatar generation.

- It costs 0.025$ per second of video, making it up 6× cheaper than similar avatar tools
- It takes 1.83s generation time per second of video, making it up to 18x faster than similar avatar tools.
- It offers full-body control, integrated audio support in 20+ languages, and is long-form video-ready for up to 3-minute consecutive clips.

We are launching with free usage until Sunday 23:59 CET at @replicate, @WiroAI, @inference_sh, @Runware, @Scenario_gg
And with general availability at:
@deepinfra, @Eachlabs

Try it now: https://t.co/AfnYID0kI2

📚 API & quickstart for Pruna API: https://t.co/fwgnfWSXIh
⭐️ Star our OSS: https://t.co/8OGBDxHpzo
🌐 Website: https://t.co/DU9Bj08Q5R

[Pruna AI @PrunaAI]

P-Video-Avatar is now the fastest model for top-quality and diverse avatar generation.

- It costs 0.025$ per second of video, making it up 6× cheaper than similar avatar tools
- It takes 1.83s generation time per second of video, making it up to 18x faster than similar avatar tools.
- It offers full-body control, integrated audio support in 20+ languages, and is long-form video-ready for up to 3-minute consecutive clips.

We are launching with free usage until Sunday 23:59 CET at @replicate, @WiroAI, @inference_sh, @Runware, @Scenario_gg
And with general availability at:
@deepinfra, @Eachlabs

Try it now: https://t.co/AfnYID0kI2

📚 API & quickstart for Pruna API: https://t.co/fwgnfWSXIh
⭐️ Star our OSS: https://t.co/8OGBDxHpzo
🌐 Website: https://t.co/DU9Bj08Q5R

[Pruna AI @PrunaAI]

First Prune is almost over 🔥

So this is our last call for anyone who wants to contribute to Pruna OSS — whether you’re already active in open source, working in ML, building in AI, using the Pruna API, or thinking about making your first OSS contribution.

If that sounds like you, there’s still time to join before April 30: pick an open issue, ask to be assigned, and submit your PR.

Each merged PR gets 60 Pruna API credits 🚀

👉 Search for your issue: https://t.co/TQJkLcsNyQ
⭐️ Star our OSS: https://t.co/8OGBDxHpzo
👉 Know more about Pruna OSS: https://t.co/JtKreCLOy7

[Pruna AI @PrunaAI]

First Prune is halfway 🚀 and we wanted to do something special:

We’re sharing a recap blog post about our OSS journey — how we started, what we’ve built so far, and what’s next.

And one more surprise... from now until April 30, each merged PR earns 60 Pruna API credits instead of 30.

If you want to join, pick an open issue, ask to be assigned, and send your PR before the end of the month.

👉 Read the blog here: https://t.co/JtKreCLOy7
👉 Search for your issue: https://t.co/TQJkLcsNyQ