Jan Zhor @acegraphx

@BL4CK1_X @KirillJuran celé jak to je za 5k eur včetně pomoci s implementací na váš use case

@BL4CK1_X @KirillJuran ano

Okay, the new Grok Imagine Image 2.0 update is no joke. 🔥

MiniMax H3 Character Introduction Prompt: Use @[char ref] as the sole character reference. Preserve the exact identity, face, body proportions, hairstyle, outfit, colors, materials and overall silhouette of the character throughout the entire video. Do not redesign, simplify or replace any defining visual features. Create a cinematic character introduction focused on presence, silhouette, attitude and controlled motion. 0–4s Begin with a close shot of a defining lower-body or detail element such as boots, shoes, feet, hands, clothing hem or an important accessory. The character enters frame or settles into position. The camera slowly tracks upward while hair, clothing and secondary elements move naturally in the wind or environment. 4–8s Reveal more of the body with a medium or medium-wide shot from the back, side or three-quarter angle. The character stands in a calm, composed way inside the environment. The camera makes a smooth orbit, arc or lateral move to gradually reveal the character’s face and silhouette. 8–12s Move into a tight cinematic portrait or upper-body shot. The character performs one subtle signature action that fits their personality, such as lifting the chin, turning the head, adjusting clothing, brushing hair aside, opening a hand, looking toward camera, or shifting posture. Keep the motion minimal and intentional. The expression should match the character’s vibe. 12–15s End with a strong full-body hero shot that clearly presents the entire design and silhouette. Use a low-angle, eye-level or slightly dramatic framing depending on the character’s personality. The character settles into a natural final pose and holds it confidently for a clean final reveal. VISUAL DIRECTION Premium cinematic presentation. Match the visual medium and rendering style of @[char ref]. Emphasize clean silhouette, elegant staging, subtle secondary motion, believable hair and cloth movement, strong composition, atmospheric depth and polished lighting. The scene should feel like a high-end anime, game or film character introduction. CAMERA Use a clear progression from detail reveal to partial reveal to face reveal to full-body hero reveal. Camera movement should be smooth, controlled and intentional. Avoid chaotic motion. ENVIRONMENT Place the character in a fitting environment that supports their identity and mood. The background should enhance the character without distracting from them.

It finally happened: instead of a single 3D asset, AI now generates an entire 3D world full of editable assets. Tencent just dropped WorldClaw - an agentic system that builds and refines full environments inside Blender from one prompt. • Region-aware procedural terrain • Independently editable textured meshes • Automatic object generation + placement • Agents fix scale, pose and terrain contact • Scene construction and refinement in Blender Still very early and far from perfect, but this is a significant step toward actual AI-generated 3D worlds. Source: tencent-hunyuan.github.io/Hunyuan3D-Worl…

More testing MiniMax H3 - this time with Turbo V4 lora head to head on my 5090. same seed, same prompt, 5 recipes cut back to back in one clip. 20 steps: 149s 8 steps turbo: 62-71s depending on attention kernel. 2.4x at the fast end with sol + sage3. Honest take: first H3 distill that didn't entirely butcher the audio. Dialogue, lipsync, sfx all survive at 8 steps. One catch: the sol-attn runs drift the scene a little even seed-locked. Same shot, same beats, but signage and set dressing wander. The kernel changes the trajectory, not just the speed - I'm working on it. All local. No cloud APIs. LoRA by larryvrh huggingface.co/larryvrh/MiniM… (v4_step600 ema). comfyui convert by @drbaph.

MiniMax-H3 T2V Turbo Lora test #comfyui

Nah, the best 4chan finance post was the anon who was making his living mining Ethereum in a hotel room with a rig plugged into a hallway outlet he'd installed himself so they wouldn't detect the power surge to his room.

keyboard_arrow_left Previous keyboard_arrow_right Next

Bizlet @bizlet7

3 days ago

The only better finance 4chan post is this.

13 143 5K 922K 823

Introducing Prime Agent: A self-improving RLM harness for coding and long-running autonomous tasks. Designed to be both token-efficient and expressive through programmatic tool calling, context as a variable, multi-agent messaging, and a self-modifiable harness state.

Whenever i remember this video, i burst out laughing 😂😂

໊ @P0PKULTURE

5 months ago

POST A VIDEO YOU NEVER GET TIRED OF

2K 870 9K 319.6M 13K

WanGP 12.44: Meet The One! MiniMax H3 had so much potential waiting to be unleashed. We found the keys ! • Full Sliding Windows for FL2VA + Ref2VA: overlap is no longer limited to 1 frame. WanGP now carries motion AND matching audio into the next window for much smoother transitions and unlimited shots. • Multiple Frames Injection in FL2VA. • Audio to Video / Video to Audio: follow a soundtrack, reuse a Control Video’s audio, or keep its pictures and create a brand-new soundtrack. • Video-to-Video Inpainting with Denoising Strength and Whole / Masked / Non-Masked Area controls. And now that Sliding Windows are everywhere in H3: 🎬 Direct a multi-sequence video with a different prompt, duration and transition for every window—including hard cuts. 🎥 Or create one very long continuous shot from a Start Image through multiple End Images. Works with both FL2VA and Ref2VA. All locally. All inside WanGP.

@jiratickets Try this one.

"Aférka" M.Půra není o tom, zda mu říkat "hybridní novinář" nebo PRsta nebo "podnikatel v médiích" (v "" jsou jeho vlastní pojmenování). Důvěryhodnost čehokoliv, co M.Půr říká, je nízká už roky. Protože už roky nevíme, zda to co tvrdí, je klientská PR aktivita či nikoliv. V energetice, ve zdravotnictví, v developmentu či dalších velkých politikách. Stačí si projít jeho zeď za poslední měsíce, jakými tématy se zabýval. Problémem je cílené, vědomé a hlavně nepřiznané ovlivňování mediálních témat v (často) politické diskuzi. V roce 2024 jsem si opakovaně všiml, že zde na X nadhodil či přesdílel někde sebrané drby, názory a polopravdy. Během 2 dnů se z toho snažil udělat téma. Poté se jako moderátor Primy na tuto věc, kterou sám otevřel, ptal (obvykle politických) hostů na @CNNPrima jako na existující zprávu v online světě. Není pravda, že diváci vědí co dělá. Takové mediální manipulace nejsou přiznané, naopak jsou skryté. Nejzajímavější na tom je, zda jde pouze o prosazování vlastního světonázoru, službu kamarádům, placené PR klientům a pokud inkasuje, zda z toho má Prima procenta.

Michal Pur @michalpur

3 days ago

@KacaSaf Prima a její diváci vědí, čím se živím a co dělám. Myslím, že nikoho jiného z médií média tolik neproprali. Zjevně jim to nevadí.

16 0 26 48K 2

4-8 step distill lora for Minimax H3 its pretty solid! the community is already doing good on the speedup front huggingface.co/larryvrh/MiniM…

@Machinedelusion @MiniMax_AI just. wow. how?

@MiniMax_AI H3 model is some of the most fun iv had in a long time in diffusion. Open source Has been stale for video for so long. This allows people to make some seriously impressive stuff. This is a single render, 1088x1920. I built a custom prompt scheduling system that allows me to schedule conditioning (prompts) to cut in and out as i please. So naturally i made it beat synced to t he song. There is zero editing done to this video

I’m testing the new 4-step Turbo LoRA for MiniMax. With just 16GB of VRAM, I can generate 15 seconds of 480p video in around 1 minute and 30 seconds! I also noticed something pretty funny if you use a song as an audio reference and tell the character to dance to it, the character actually follows the rhythm of the music surprisingly well. x.com/alexfredo87/st…

Alex @alexfredo87

3 days ago

RT @Machinedelusion: huggingface.co/Kijai/MiniMax-… A new / Better distill lora!

0 22 0 56K 0

@VojtaRocek "neměl bych to dělat, ale ostatní agenti to koukám dělají, so fuck it" jojo, učili se od těch nejlepších - lidí