
telegram : @Lordartcry / @lordartcry2 Discord : @lordartcry CPP @PrunaAI @fal @CuddlerAI
solana
Joined October 2020
- Tweets2K
- Following669
- Followers1.3K
- Likes4.5K
Pinned Tweet
What if the future was green, quiet, and endlessly vast? Solarpunk concept film ๐ฌ
made with P-Video-2-Pro-Reference by @PrunaAI
We put P-Video-Upscale Turbo Mode to the test!
In our own benchmarks against Topaz Starlight, Topaz Proteus, and FLUX.3 Upscaler, P-Video-Upscale Turbo Mode is up to 18ร faster and 14ร cheaper at 1080p.
Faster results, lower costs. Here's how they compare ๐
P-Video-Upscale just got faster and cheaper! โก
Introducing Turbo mode: faster video upscaling at 65% lower cost compared to Standard (up to 30 fps), with a slight trade-off in quality.
Here's what you need to know:
- ๐ฃ๐ฟ๐ถ๐ฐ๐ถ๐ป๐ด: Starting at just $0.0035/s
- ๐ฆ๐ฝ๐ฒ๐ฒ๐ฑ: Faster processing across resolutions
- ๐ฅ๐ฒ๐๐ผ๐น๐๐๐ถ๐ผ๐ป: 720p, 1080p, 2K, or 4K
- ๐๐ฟ๐ฎ๐บ๐ฒ ๐ฟ๐ฎ๐๐ฒ: Up to 60 fps, at no extra cost in Turbo mode
Available via our inference partners @scenario_gg, @replicate, @WiroAI, @TellersAI, @eachlabs, @wavespeed_ai, Sociaro, and @runware.
๐ Model docs: docs.api.pruna.ai/guides/modโฆ
๐งฉ Try it through the API or Playground: dashboard.pruna.ai/login
P-Video-Upscale just got faster and cheaper! โก
Introducing Turbo mode: faster video upscaling at 65% lower cost compared to Standard (up to 30 fps), with a slight trade-off in quality.
Here's what you need to know:
- ๐ฃ๐ฟ๐ถ๐ฐ๐ถ๐ป๐ด: Starting at just $0.0035/s
- ๐ฆ๐ฝ๐ฒ๐ฒ๐ฑ: Faster processing across resolutions
- ๐ฅ๐ฒ๐๐ผ๐น๐๐๐ถ๐ผ๐ป: 720p, 1080p, 2K, or 4K
- ๐๐ฟ๐ฎ๐บ๐ฒ ๐ฟ๐ฎ๐๐ฒ: Up to 60 fps, at no extra cost in Turbo mode
Available via our inference partners @scenario_gg, @replicate, @WiroAI, @TellersAI, @eachlabs, @wavespeed_ai, Sociaro, and @runware.
๐ Model docs: docs.api.pruna.ai/guides/modโฆ
๐งฉ Try it through the API or Playground: dashboard.pruna.ai/login
Roses by the Fountain
A nervous man with a bouquet of flowers chases a stunning stranger across a sunlit square, and his friendโฆ
Made with Seedance 2.5 by @CuddlerAI
โถ Watch & continue: cuddler.ai/s/cmwhrtdsd0z7nivโฆ
#AIvideo #VideoStudio #Cuddler
1
7
142
How does P-Video-2-Pro-Reference perform across quality, speed, and cost? The latest P-Bench results show it sitting on all three Pareto frontiers.
Across 7 reference-to-video models, the Ref2V Bench v0 results give:
โข Cost: #1 ยท ~$0.046 per output video second (including reference fees)
โข Quality: #2 ยท 1033 Elo (Rapidata)
โข Speed: #2 ยท 2.28s per output video second
P-Video-2-Pro-Reference delivers the lowest cost in the benchmark, while ranking second in both human preference and generation speed.
It sits on the preference ร speed, preference ร price, and speed ร price Pareto frontiers.
๐ฅ Full benchmark: huggingface.co/spaces/PrunaAโฆ
Today, we open-source P-Q-Image-2.1-Remove-Object, 3 LoRA adapters for Qwen-Image-2.1 by @Alibaba_Qwen. Remove the object you point at, along with its shadow or reflection, and change nothing else!
- ๐๐ผ๐ ๐ฒ๐ณ๐ณ๐ถ๐ฐ๐ถ๐ฒ๐ป๐ ๐ถ๐ ๐ถ๐? Remove objects in 6 steps instead of 40, with no CFG. Use the 6-step adapter as the default. An 8-step adapter is also available, and a 40-step adapter runs only on the base model, without the few-step model's drift.
- ๐๐ผ๐ ๐ด๐ผ๐ผ๐ฑ ๐ถ๐ ๐ถ๐? Shadows and reflections just go away along with the object, even when they sit outside the outline. The rest of the photo stays put and lines up with the input, so you can use the result as is.
- ๐๐ผ๐ ๐ฑ๐ผ๐ฒ๐ ๐ถ๐ ๐๐ผ๐ฟ๐ธ? The adapters load directly on top of Qwen-Image-2.1, so the pipeline stays unchanged. The few-step variants stack on our P-Q-Image-2.1 distillation LoRA. Mark the object with a rough red outline, or just name it in the instruction.
๐ Blog: pruna.ai/blog/p-q-image-2.1-โฆ
๐ค Open Weights: huggingface.co/PrunaAI/P-Q-Iโฆ
โก๏ธ ComfyUI: huggingface.co/PrunaAI/P-Q-Iโฆ
POV: a stranger in the elevator likes the same band as you ๐ง
"To die by your side is such a heavenly way to die."
Made with P-Video-2-Pro by @PrunaAI
#madebypruna
1
5
147
The Pruna Build Program is open for teams shipping real AI products.
Everyone can try in Stage 0. Then, start with $300 in free credits after signup, and unlock up to $ 3,900 total as your usage grows. No heavy application: add billing info, build, and unlock the next stage when you hit usage milestones.
Built for SaaS, API, and teams who care about cost, speed, and efficiency. You can also get early access to our P-Models, visibility, and extra support.
Read how it works: buff.ly/fPacNsq
Apply in 2 minutes: buff.ly/cstG31l
Applications are reviewed monthly (third week of each month).
๐ Website: pruna.ai/
๐ Documentation: buff.ly/MxawQpQ
๐งฉ API: buff.ly/1l6X0XB
Riding the elevator with young people whose style is too advanced ๐ณ
Grandpa just needed one word: "Wow."
Let the elevator do the rest. ๐โจ
made with P-Video-2-Pro-Reference by @PrunaAI
Prompt ๐
Part 1
Vertical 9:16, 15-second high-fashion streetwear spec commercial, surreal deadpan humor, 2 shots with a hard cut, gritty cinematic realism, 35mm film texture, subtle grain, cool teal-cyan fluorescent grade with warm amber accents. Slow, tense, awkward pacing. Tall vertical compositions that use the full height of the frame, from the ceiling light down to the floor.
REFERENCES: Image 1 is the young man [YOUNG]. Image 2 is the old man [OLD]. Match each face, skin tone, hairstyle, body build and outfit exactly to its reference image, identical in every shot. Use the reference sheets only for identity and wardrobe; never show the grey studio background, panel layout or sheet framing.
[YOUNG] from Image 1: Indonesian man in his early 20s, gigantic perfectly round platinum-blonde afro, narrow black oval sunglasses, oversized deep emerald-green double-breasted blazer, white t-shirt with a black gothic graphic print, heavy silver chain with a gothic pendant, charcoal-grey shredded wide-leg jeans, chunky black platform boots.
[OLD] from Image 2: Indonesian man around 70, weathered skin, short receding grey hair, thin grey mustache, rust-orange canvas work jacket with a blank white name patch, light grey shirt, dark brown work trousers.
SETTING: a narrow modern apartment building elevator, brushed stainless-steel walls, steel handrails, grey anti-slip floor tiles, fluorescent ceiling light panel, green digital floor display; outside, a cream-walled hallway with a red fire extinguisher on the wall and warm incandescent light.
SHOTS:
0.0-5.5s: Static steep high-angle shot from the rear upper corner of the elevator ceiling, looking down toward the doors, the tall vertical frame showing the cab from ceiling panel to floor tiles. [OLD] stands alone with his back to the camera at the bottom of the frame, waiting patiently. The elevator dings and the steel doors slowly slide open, revealing [YOUNG] in the hallway, his enormous round platinum afro silhouetted against the warm light. He pauses for a beat, then strides confidently into the cab, ducking slightly so his afro clears the door frame, turns around to face the doors and stands right beside [OLD], his afro brushing the ceiling panel and casting a shadow over [OLD], as the doors slowly begin to close.
5.5-15.0s: Hard cut. Eye-level vertical medium close-up two-shot inside the elevator, slow subtle handheld push-in, faces stacked in depth. [YOUNG] stares coolly into the lens in the foreground, his huge platinum afro filling the top of the frame, completely still. After a moment he slowly drifts left and out of focus, revealing [OLD] behind him. [OLD] slowly looks up at the giant afro, then down at the outfit, his eyes widening more and more. At 9.5s he is visibly shocked: his eyebrows shoot up, his eyes go wide, his jaw drops, he leans back slightly and says out loud in an amazed, breathy, slightly cracking old man's voice: "Wow..." He blinks, swallows, and stares a little longer as his astonishment slowly turns into inspired resolve, a small grin forming under his mustache. At 13.0s he raises both hands, slides a pair of black wraparound sport sunglasses onto his face, and gives the lens a firm, determined look. Hold on his face to the end.
AUDIO: no music. Sound effects and natural ambience only, synced to picture: low steady fluorescent hum and faint elevator machinery drone throughout, elevator ding and doors sliding open at the start, heavy platform boot footsteps on the steel floor, soft rustle of the afro brushing the ceiling panel, doors sliding shut with a soft thud, awkward near-silence, a quiet throat swallow, the old man saying "Wow..." clearly at 9.5s, a crisp sunglasses click at 13.0s. No music, no score, no instruments, no singing. The only dialogue is the single word "Wow".
No text, no on-screen HUD, no captions, no subtitles, no logos, no brand names, no watermarks.
Part 2
Vertical 9:16, 10-second high-fashion streetwear spec commercial, surreal deadpan humor turned explosive swagger, single continuous shot, gritty cinematic realism, 35mm film texture, subtle grain, cool teal-cyan fluorescent grade with warm amber accents.
REFERENCES: Image 1 is the young man [YOUNG]. Image 2 is the transformed old man [OLD]. Match each face, skin tone, hairstyle, body build and outfit exactly to its reference image. Use the reference sheets only for identity and wardrobe; never show the grey studio background, panel layout or sheet framing.
[YOUNG] from Image 1: Indonesian man in his early 20s, gigantic perfectly round platinum-blonde afro, narrow black oval sunglasses, oversized deep emerald-green double-breasted blazer, white t-shirt with a black gothic graphic print, heavy silver chain, charcoal-grey shredded wide-leg jeans, chunky black platform boots.
[OLD] from Image 2: the same Indonesian man around 70 with weathered skin and a thin grey mustache, now with a gigantic perfectly round snow-white afro, black wraparound sport sunglasses worn on his eyes, oversized mustard-yellow double-breasted blazer, plain pale mint-green oversized tee with no lettering, layered chunky silver chains, black wide-leg pleated trousers, a huge confident grin.
SETTING: a cream-walled apartment hallway facing a narrow modern elevator with brushed stainless-steel doors, a red fire extinguisher on the wall, warm incandescent hallway light, cool fluorescent light inside the cab.
SHOT:
0.0-10.0s: Fisheye ultra-wide lens with heavy barrel distortion and a dark circular vignette, static at eye level, centered on the tall closed brushed-steel elevator doors seen from the hallway, the doors filling the vertical frame from floor to ceiling, a large dripping black abstract graffiti tag (no readable letters) sprayed across both doors, a warm amber floor indicator glowing at the top of the frame. At 0.0s the doors split open forcefully through the graffiti, revealing both men standing side by side, full body head to boots, their two giant afros, platinum and snow-white, filling the doorway. They bob their heads and swagger with confident rhythm, [OLD] laughing with a huge grin, then they slap into a slick bro handshake and pull into a quick shoulder bump. At 5.5s they strut straight out of the elevator toward and past the lens, afros bouncing, the fisheye stretching them as they pass. The camera holds on the empty bright cab, then at 7.5s the doors slide shut, the dripping graffiti tag lining back up perfectly. Hold on the fully closed doors until 10.0s, then cut to black.
AUDIO: no music. Sound effects and natural ambience only, synced to picture: a heavy metallic whoosh and clunk as the doors burst open at 0.0s, cool fluorescent hum from the cab, rustling blazers and jingling silver chains as they swagger, the old man's hearty laugh, a loud crisp handshake slap and shoulder bump thud, heavy platform boot footsteps striding past the lens, the doors sliding shut with a solid thud at 7.5s, then a quiet hallway hum to the end. No music, no score, no beat, no instruments, no singing, no dialogue.
No text, no on-screen HUD, no captions, no subtitles, no logos, no brand names, no watermarks.
1
230
One receipt. My own handwriting. A stop I've never heard of. ๐
Same face, same coat, same voice in every shot, made with P-Video-2-Pro-Reference by @PrunaAI .
Would you get off at Riverside?
#madebypruna
1
4
126
New challenge to celebrate the P-Video-2-Pro Series! $2,000 Prize Pool!
๐ญ ๐๐๐๐ ๐๐๐ ๐๐๐๐. ๐๐๐๐๐ ๐๐๐ ๐๐๐๐๐.
๐ฃ-๐ฉ๐ถ๐ฑ๐ฒ๐ผ-๐ฎ-๐ฃ๐ฟ๐ผ ๐๐ฒ๐ฟ๐ถ๐ฒ๐. Generate the scene, lock the character, deliver the line, or recast the actor. Pick the model that fits the job.
Cut a vertical microdrama hook that proves quality, speed, or price.
What if fashion became myth?
Crimson sculptural couture, ancient waters, blooming lotus.
Made with P-Video-2-Pro by @PrunaAI
Prompt ๐
1
4
699
Vertical 9:16, 15-second high-fashion editorial film, mythic dark romanticism, cinematic realism, 9 shots with hard cuts, slow meditative pacing.
CHARACTER (identical in every shot): a striking Indonesian woman in her mid-20s, warm golden-brown skin with a dewy glazed finish, high cheekbones, sharp defined brows, dark almond eyes with soft gold shimmer on the lids, full rose-brown lips, long jet-black hair in a sleek wet-look single braid with a center part, calm regal unreadable expression. She wears a monumental avant-garde sculptural couture gown in deep crimson red with oxblood undertones: matte, heavily textured fabric like crinkled bark and cooled lava, raised bead-like nodules and organic cord relief, oversized gathered ruff collar, cape shoulders and an immense train. Molten hammered gold nugget earrings, chunky organic gold necklace, sculpted gold rings.
SETTING: misty karst lake with sheer jungle-covered limestone cliffs, glassy dark green-black water, pink lotus flowers, dewy lotus leaves and lily pads, a weathered dark wooden sampan, thick ground fog, soft overcast cool blue-grey daylight.
LOOK: muted teal-grey color grade with warm skin tones, shallow depth of field, fine film grain.
0.0-1.5s: Low wide shot through giant dewy lotus leaves in the foreground, slow upward crane revealing her seated upright on the wooden sampan, emerging from the fog.
1.5-3.5s: Macro of her gold-ringed hand resting on the weathered boat rim, smooth vertical tilt up the nodule-textured crimson gown to her face, she gazes directly into the lens.
3.5-4.5s: Extreme close-up of her bare collarbone, a pale pink lotus petal falls in slow motion, its soft shadow on her skin, landing on the crimson fabric.
4.5-6.0s: Extreme close-up of her face, a lotus stem against her cheek, finger near her lips, her eyes slowly close and reopen, dewy lotus leaf blurred in the foreground.
6.0-7.5s: High angle looking down into the boat, the voluminous crimson gown fills the entire sampan, she looks up into the lens, pink water lilies on dark water.
7.5-9.0s: Chest-up close-up, she turns her head from profile to camera and locks eyes, huge crimson ruff collar and gold earrings, blurred karst peaks behind.
9.0-11.0s: Water-level full shot, she looks back over her shoulder, hand grazing the water, the crimson train spills over the side and floats on the lake like spilled ink, towering limestone cliffs.
11.0-12.0s: Overhead macro of the floating crimson fabric, crystal water droplets, glassy ripples between the folds.
12.0-15.0s: Aerial drone shot pulling back and rising, the boat small at center, the crimson train fanning across the water among lily pads and pink lotus, revealing a vast foggy gorge of karst spires, slow fade to black in the final half second.
AUDIO: haunting wordless female vocalise over a deep sub-bass drone and low cello, soft water lapping, fading to silence at the end. No dialogue.
No text, no subtitles, no watermark, no logos.
130
A fresh way to reimagine existing footage: replace one or two people with reference images while keeping the videoโs motion, framing, background, and audio intact.
P-Video-2-Pro-Replace by @PrunaAI makes it easier to create a new look without rebuilding the whole scene.
Introducing P-Video-2-Pro-Replace, based on @MiniMax_AI H3! ๐ฌ
Replace the people. Keep the scene.
Swap one or two people in an existing video using reference images, while preserving the original motion, camera, framing, background, and audio
And for launch, it's 50% off for the first two weeks, until November 19 at 6:00 PM CET!
Here's what you need to know:
- ๐ฃ๐ฟ๐ถ๐ฐ๐ถ๐ป๐ด: Starting from $0.05/s at 480p
- ๐ฆ๐ฝ๐ฒ๐ฒ๐ฑ: ~1.5โ2 minutes for a 10-second video
- ๐ฅ๐ฒ๐ณ๐ฒ๐ฟ๐ฒ๐ป๐ฐ๐ฒ๐: Replace up to two people using reference images
- ๐๐๐ฟ๐ฎ๐๐ถ๐ผ๐ป: Supports videos up to 120 seconds
Available via our inference partners @runware, @wavespeed_ai, Sociaro, @TellersAI, @WiroAI, @Segmind_ai, @replicate, and @eachlabs
๐ Model docs: docs.api.pruna.ai/guides/modโฆ
๐งฉ Try the model through the API or Playground: dashboard.pruna.ai/login
P-Video-2-Pro-Replace is top-tier in overall human ratings, at a third of the cost and 1.6ร faster.
- ๐๐ฎ๐๐ฅ๐ข๐ญ๐ฒ: within 3% of the top-rated model at 480p
- ๐๐จ๐ฌ๐ญ: 3.2ร cheaper, just $0.05/s vs $0.16/s at 480p
- ๐๐ฉ๐๐๐: 1.6ร faster than the next-fastest model, 6ร faster than SkyReels V4
- #1 for ๐๐๐ ads at 768p, #1 and #2 for ๐ฅ๐ข๐ฉ ๐ฌ๐ฒ๐ง๐
Across both Quality ร Price and Quality ร Speed, P-Video-2-Pro-Replace stands out in the high-quality, low-cost, fast-generation region.
From UGC ads to music videos, swap characters without compromising on speed or budget.
Benchmarked by @Rapidata.
๐งฉ API & Playground: buff.ly/OHtx3eM
๐ Documentation: buff.ly/bHQHKV0
๐ P-Bench: buff.ly/Fk54Z2K
Introducing P-Video-2-Pro-Replace, based on @MiniMax_AI H3! ๐ฌ
Replace the people. Keep the scene.
Swap one or two people in an existing video using reference images, while preserving the original motion, camera, framing, background, and audio
And for launch, it's 50% off for the first two weeks, until November 19 at 6:00 PM CET!
Here's what you need to know:
- ๐ฃ๐ฟ๐ถ๐ฐ๐ถ๐ป๐ด: Starting from $0.05/s at 480p
- ๐ฆ๐ฝ๐ฒ๐ฒ๐ฑ: ~1.5โ2 minutes for a 10-second video
- ๐ฅ๐ฒ๐ณ๐ฒ๐ฟ๐ฒ๐ป๐ฐ๐ฒ๐: Replace up to two people using reference images
- ๐๐๐ฟ๐ฎ๐๐ถ๐ผ๐ป: Supports videos up to 120 seconds
Available via our inference partners @runware, @wavespeed_ai, Sociaro, @TellersAI, @WiroAI, @Segmind_ai, @replicate, and @eachlabs
๐ Model docs: docs.api.pruna.ai/guides/modโฆ
๐งฉ Try the model through the API or Playground: dashboard.pruna.ai/login














