High-performance AI models that combine speed, quality, and specialization for key use cases. ๐ฑ Make AI accessible and sustainable for everyone!
Munich & Paris
- Tweets 1,011
- Following 224
- Followers 4,539
- Likes 2,208
Today, we launch P-Video-2-Pro, based on MiniMax H3: The best quality. The best speed.
Text or image in. Video and audio out.
โข ๐๐ป๐ฝ๐๐: Text or first-frame image, with optional last-frame conditioning
โข ๐ฆ๐ฝ๐ฒ๐ฒ๐ฑ: ~2.0s for a 5s clip at 480p ยท ~4.3s at 768p in Speed mode
โข ๐ฃ๐ฟ๐ถ๐ฐ๐ถ๐ป๐ด: From $0.02/s Speed ยท $0.04/s Quality
โข ๐ ๐ผ๐ฑ๐ฒ๐: Speed or Quality
โข ๐๐ผ๐ป๐๐ฟ๐ผ๐น: First and last frame, seed, aspect ratio, and prompt upsampling with Off, Turbo, or Max
Available via our inference partners @replicate, @Cloudflare, @eachlabs, @scenario_gg, @TellersAI, @wavespeed_ai, @wiroai, @runware, and @togethercompute
Validated by our benchmark partners @datapointai, @RapidataAI, and @DesignArena. Benchmarks to be released soon ๐
๐ Free Playground: playground.pruna.ai/
๐ Model docs: docs.api.pruna.ai/guides/modโฆ
๐งฉ API: dashboard.pruna.ai/login
37
57
349
100,737
The Pruna-Qwen-Image-2.1 Space is now live!
You can now test our few-step LoRA adapters for Qwen-Image-2.1 by @Alibaba_Qwen directly in your browser, no setup or code required.
- 5 steps for maximum speed
- 8 steps for the best balance of speed and quality
- Text-to-image and image editing support
Put it to the test and show us the results. B๐ฆ๐ต๐ต๐ฆ๐ณ ๐ท๐ฆ๐ณ๐ด๐ช๐ฐ๐ฏ๐ด ๐ฐ๐ฏ ๐ต๐ฉ๐ฆ ๐ธ๐ข๐บ. ๐
Test the Pruna-Qwen-Image-2.1 Space on @huggingface by @_akhaliq: huggingface.co/spaces/PrunaAโฆ
Check Pruna-Qwen-Image-2.1: huggingface.co/PrunaAI/Prunaโฆ
6
16
99
7,577
Pruna AI retweeted
We're open sourcing ultra fast loras for Qwen-Image-2.1!
Check them out here and let us know how we can improve them!
huggingface.co/PrunaAI/Prunaโฆ
Today, we open-source Pruna-Qwen-Image-2.1, a set of a few-step LoRA adapters that make Qwen-Image-2.1 by @Alibaba_Qwen up to 6.3ร faster for image generation and editing.
- ๐๐ผ๐ ๐ฒ๐ณ๐ณ๐ถ๐ฐ๐ถ๐ฒ๐ป๐ ๐ถ๐ ๐ถ๐? Generate or edit images in just 5 or 8 steps instead of 40. Choose 5 steps for maximum speed or 8 steps for the best balance.
- ๐๐ผ๐ ๐ด๐ผ๐ผ๐ฑ ๐ถ๐ ๐ถ๐? The 8-step adapter is our recommended default. It supports text-to-image and single- or multi-image editing at 1K resolution, with up to three reference images.
- ๐๐ผ๐ ๐ฑ๐ผ๐ฒ๐ ๐ถ๐ ๐๐ผ๐ฟ๐ธ?The LoRA adapters load directly on top of Qwen-Image-2.1, while the pipeline remains unchanged. Each adapter uses its own optimized sigma schedule and runs without CFG.
๐๐ฉ๐ช๐ด ๐ช๐ด ๐ซ๐ถ๐ด๐ต ๐ฐ๐ถ๐ณ ๐ง๐ช๐ณ๐ด๐ต ๐ณ๐ฆ๐ญ๐ฆ๐ข๐ด๐ฆ, ๐ธ๐ช๐ต๐ฉ ๐ฆ๐ท๐ฆ๐ฏ ๐ฃ๐ฆ๐ต๐ต๐ฆ๐ณ ๐ท๐ฆ๐ณ๐ด๐ช๐ฐ๐ฏ๐ด ๐ฐ๐ฏ ๐ต๐ฉ๐ฆ ๐ธ๐ข๐บ. ๐
Want to try the fast open-source integration? Check it here on Diffusers: buff.ly/1CpiRPp
Want to try the fastest image generation and editing endpoint? Check P-Image-Ideogram and P-Image-Edit here on API: buff.ly/gw5tvUV
3
13
129
12,432
Super proud to announce that @PrunaAI is releasing a new open-source Lora for Qwen-Image-2.1.
Amazing aesthetics and super fast.
Thank you @Alibaba_Qwen for this amazing model for the community
Made with AI
3
6
236
Pruna AI retweeted
I'm grateful for the open ecosystem around generative models I get to work in. Today it's time to give a little back.
We're releasing step-distillation LoRAs for Qwen-Image-2.1.
Generate and edit images in 5 or 8 steps, up to 6.3x faster.
1
4
11
293
Today, we open-source Pruna-Qwen-Image-2.1, a set of a few-step LoRA adapters that make Qwen-Image-2.1 by @Alibaba_Qwen up to 6.3ร faster for image generation and editing.
- ๐๐ผ๐ ๐ฒ๐ณ๐ณ๐ถ๐ฐ๐ถ๐ฒ๐ป๐ ๐ถ๐ ๐ถ๐? Generate or edit images in just 5 or 8 steps instead of 40. Choose 5 steps for maximum speed or 8 steps for the best balance.
- ๐๐ผ๐ ๐ด๐ผ๐ผ๐ฑ ๐ถ๐ ๐ถ๐? The 8-step adapter is our recommended default. It supports text-to-image and single- or multi-image editing at 1K resolution, with up to three reference images.
- ๐๐ผ๐ ๐ฑ๐ผ๐ฒ๐ ๐ถ๐ ๐๐ผ๐ฟ๐ธ?The LoRA adapters load directly on top of Qwen-Image-2.1, while the pipeline remains unchanged. Each adapter uses its own optimized sigma schedule and runs without CFG.
๐๐ฉ๐ช๐ด ๐ช๐ด ๐ซ๐ถ๐ด๐ต ๐ฐ๐ถ๐ณ ๐ง๐ช๐ณ๐ด๐ต ๐ณ๐ฆ๐ญ๐ฆ๐ข๐ด๐ฆ, ๐ธ๐ช๐ต๐ฉ ๐ฆ๐ท๐ฆ๐ฏ ๐ฃ๐ฆ๐ต๐ต๐ฆ๐ณ ๐ท๐ฆ๐ณ๐ด๐ช๐ฐ๐ฏ๐ด ๐ฐ๐ฏ ๐ต๐ฉ๐ฆ ๐ธ๐ข๐บ. ๐
Want to try the fast open-source integration? Check it here on Diffusers: buff.ly/1CpiRPp
Want to try the fastest image generation and editing endpoint? Check P-Image-Ideogram and P-Image-Edit here on API: buff.ly/gw5tvUV
31
100
929
81,801
This week we're reviewing your ๐ฃ๐ฟ๐๐ป๐ฎ ๐๐ฟ๐ฒ๐ฎ๐๐ถ๐๐ฒ ๐ฃ๐ฟ๐ผ๐ด๐ฟ๐ฎ๐บ โจ applications! Hurry up and apply now to be considered!
- ๐ช๐ต๐ฎ๐โ๐ ๐ถ๐ ๐ฎ๐ฏ๐ผ๐๐? A space for AI artists and creators to experiment with faster, more affordable, and controllable models.
- ๐ช๐ต๐ผโ๐ ๐ถ๐ ๐ณ๐ผ๐ฟ? The program is open to artists, filmmakers, animators, designers, developers, creative technologists, and AI-native creators.
- ๐ช๐ต๐ฎ๐ ๐ฑ๐ผ ๐๐ผ๐ ๐ด๐ฒ๐? Become an ๐๐
๐ฝ๐น๐ผ๐ฟ๐ฒ๐ฟ or a ๐๐ฒ๐น๐น๐ผ๐ and get supported with monthly credits, early model access, rewarded campaigns, and opportunities to be featured.
No need for thousands of followers. Just original ideas.
Your work. Your style. Your creative freedom.
Read more information and apply here: buff.ly/9xNoyxr
9
7
42
1,882
Pruna AI retweeted
Tried pushing Pruna P-video-2-pro into something more surreal with this one . Realistic Seoul slowly collapsing into an origami world.
The way it handles large environmental transformations and cinematic motion made this concept way more fun to build.
@PrunaAI #madebypruna
2
46
9,337
The clothes are waiting. One choice. One look. One transformation.
Product P-Video-2-Pro generates a cinematic video from text, a first-frame image, or the first and last frames. Picture and sound in one clip, up to 15 seconds.
Speed from $0.02/s of output video ยท Quality from $0.04/s
What's the setup? ๐
- Model: P-Video-2-Pro
- Settings: Quality mode, 768p, prompt_upsampling max
- Input prompt in the comments ๐
๐งช Free Playground: buff.ly/SFnS9yZ
๐ Documentation: buff.ly/0P6Derc
๐งฉ API: buff.ly/1l6X0XB
6
9
39
2,458
Prompt: Cinematic high-fashion commercial. The video opens with a circular arrangement of stylish garments hanging individually from elegant clothes hangers, suspended in the space like a fashion installation. The clothes slowly rotate in a circle while the camera moves around them, revealing different textures, silhouettes, and fabrics. Cut to an extreme close-up of two selected garments. The camera glides slowly across the fabric, stitching, folds, and details. A stylish model enters the frame, walking confidently. The camera subtly tracks the model. The model stops in front of the rack, looks through the clothes, and selects one garment with their hand. The garment disappears from their hand and instantly appears perfectly dressed on their body. The model pauses, looks at the new outfit, then confidently turns toward the camera. The model begins posing naturally and confidently in the outfit: strong editorial poses, slight turns of the body, subtle movement of the fabric, and direct looks toward the camera.
5
695
Pruna AI retweeted
Just wrapped a 30-second animated story featuring a little pirate battling storms for treasure. Used P-Video-2-Pro by @PrunaAI, and honestly, it nailed the character consistency. Shifting from a chaotic ocean scene to a glowing treasure cave was a real test for AI animation. But it handled dynamic lighting and character features like a champ, making it feel like a true 3D short film. Always amazed by what AI can pull off in animation these days.
#madebypruna #PVideo2Pro #AIanimation #CinematicAI
Made with AI
Paid partnership (ad)
21
2
33
37,764
the storm was loud. the light was louder.
made this 30s cinematic story with P-Video-2-Pro, turning a lighthouse, a distant boat and a storm into a little story about finding the way home.
really liked the cinematic camera movement and the way the light cuts through the rain.
@PrunaAI #madebypruna
9
1
23
3,159
Pruna AI retweeted
A puppy, a kitten, and one flower that started a friendship. ๐ถ๐ฑ๐ธ A whimsical 30-second animated short made with @PrunaAI's P-Video-2-Pro: cinematic visuals and native audio, glowing fireflies included.
1
1
4
9,283
Pruna AI retweeted
๋น ์ค๋ ๋ฐค, ์ฐ์ฐํ ํ๋์์ ๋จ์ด์ง ์์ ๋ณ์ ๋ฐ๊ฒฌํ๋ค.
๊ทธ ๋ณ์ ๋ค์ ํ๋๋ก ๋๋ ค๋ณด๋ด๊ธฐ ์ํด ์ฅ์๊น์ง ๋ฌ๋ ค๊ฐ๋ ์ด์ผ๊ธฐ๋ฅผ 30์ด ์ ๋๋ฉ์ด์
์ผ๋ก ๋ง๋ค์ด๋ดค์ด์.
P-Video-2-Pro์ ์๋ค๋งํฑํ ์ฐ์ถ๊ณผ ๋น, ๋น, ์นด๋ฉ๋ผ ์์ง์์ ํ์ฉํด ์์ ์ด์ผ๊ธฐ๋ฅผ ๋ด์๋ดค์ต๋๋ค.
@PrunaAI #madebypruna
6
3
23
3,166
Pruna AI retweeted
when the world falls apart, sometimes saving one life is enough to change everything.
i created this story with P-Video-2-Pro, and iโm honestly impressed by how @PrunaAI brings cinematic worlds, expressive characters, detailed environments, and dynamic camera movement to life.
inspired by the feeling of trust and friendship in dragon stories, but with my own characters and story.
@PrunaAI #madebypruna
7
1
16
3,680
New challenge to celebrate P-Video-2-Pro! $2,000 Prize Pool
๐จ ๐๐๐๐ ๐๐๐๐๐ ๐๐ ๐๐๐๐
๐ฃ-๐ฉ๐ถ๐ฑ๐ฒ๐ผ-๐ฎ-๐ฃ๐ฟ๐ผ, based on MiniMax H3. Quality-tier video generation model. First and Last Frame. 480p/768p with generated audio. Speed and Quality mode. Off, turbo, or max prompt upsampler.
Your story starts here.
18
26
107
7,464
How to enter:
1. Create a 30-second video that tells a short story using P-Video-2-Pro. Tell us your story, a known story, a new one, or someone's story. Be creative and original.
2. Post on X or Instagram
3. Tag us and add #madebypruna
4. Submit your post link on @GimiApp: app.gimi.co/campaigns/your-sโฆ
Use other tools or P-models to improve the final content quality, but donโt use them to create misleading or deceptive content.
Join our Discord to discuss and share about it: discord.com/invite/JFQmtFKCjโฆ
1
7
705
Available until rewards run out. Donโt miss it!
Start creating now:
โข API (~2.5K free generations after creating the profile): dashboard.pruna.ai/login
โข Playground (5 free generations, back again soon for P-Video-2-Pro): playground.pruna.ai
1
6
595
Nothing is more satisfying than P-Video-2-Pro. Breathe and relax.
P-Video-2-Pro generates a cinematic video from text, a first-frame image, or the first and last frames. Picture and sound in one clip, up to 15 seconds.
Speed from $0.02/s of output video ยท Quality from $0.04/s
๐งช Free Playground: buff.ly/SFnS9yZ
๐ Documentation: buff.ly/0P6Derc
๐งฉ API: buff.ly/1l6X0XB
8
12
70
5,306
