π“Ÿ host1.bitchan.one

𓀐 @ikioJ-Fx

⏱ 1w seenβ›“ 0πŸ‘ 0β§– 2

filter [-]

languages:onother

>mfw Resource news 10/30/2026 >Pika H3 Flex: Built for professional iteration https://huggingface.co/Pika-Labs/Pika-H3-Flex >H3 Motion: Lift a subject out of a photo and drag it into motion https://github.com/PxTicks/h3-motion >Idol: A single image to a 3D Gaussian splat https://huggingface.co/speridlabs/idol https://github.com/Nynxz/ComfyUI-Idol >Don't Ask the LLM: Visual Token Pruning with Spectral Structure and Semantic Relevance https://github.com/sejong-rcv/ST-Pruner >MEV: A Multi-Event Vi…⏱ 1w↩
>mfw Resource news 10/30/2026 >Pika H3 Flex: Built for professional iteration https://huggingface.co/Pika-Labs/Pika-H3-Flex >H3 Motion: Lift a subject out of a photo and drag it into motion https://github.com/PxTicks/h3-motion >Idol: A single image to a 3D Gaussian splat https://huggingface.co/speridlabs/idol https://github.com/Nynxz/ComfyUI-Idol >Don't Ask the LLM: Visual Token Pruning with Spectral Structure and Semantic Relevance https://github.com/sejong-rcv/ST-Pruner >MEV: A Multi-Event Video Dataset for Long-Take Generation https://mev-dataset.github.io >IntReal: A Real-World Dataset and Benchmark for Multimodal Interleaved Generation https://yushaohan.github.io/IntReal >Wavelet Phase Diffusion for Structurally and Semantically Consistent Sim-to-Real Translation https://kit-mrt.github.io/Wavelet-Phase-Diffusion >CreatorOS Studio: One recurring character, from a one-line brief to a finished reel https://github.com/thePloggy/CreatorOS-Studio >ZenTrainer: Train LoRAs inside ComfyUI, and watch them learn https://github.com/Nynxz/ZenTrainer 10/29/2026 >NVIDIA FastGen: Fast Generation from Diffusion Models https://github.com/NVlabs/fastgen >minimax_h3_open_regenerate_2k_refiner_rank_384_bf16 https://huggingface.co/Kijai/MiniMax-H3-experimental/blob/main/loras/minimax_h3_open_regenerate_2k_refiner_rank_384_bf16.safetensors >SpEdE β€” Splat Edit Engine for ComfyUI https://github.com/Burgstall-labs/ComfyUI-SpEdE >Directors Cut: A local asset and REF2VA video studio for ComfyUI, powered by Qwen-Image 2.1, MiniMax H3 https://github.com/leonardofrito/Directors-Cut >AReA: Training-free attention mechanism for generating images beyond DiT training resolutions https://github.com/KooroshRH/AReA >Hy-Nomas: Towards Free-Form Interactive World Generation https://hy-nomas.hy.tencent.com 10/28/2026 >Open-H3-Regenerate-2K: High-Resolution Video Refinement with Few-Step Diffusion https://nvlabs.github.io/Sana/Open-H3-Regenerate-2K⏱ 1w↩ 1
Thank you, I need to read the news to get hard and gen adequately.⏱ 1w↩