Repackaged model files for ComfyUI. - https://huggingface.co/nvidia/ChronoEdit-14B-Diffusers - https://huggingface.co/Wan-AI/Wan2.2-Animate-14B - https://huggingface.co/alibaba-pai/Wan2.2-Fun-A14B-Control-Camera - https://huggingface.co/alibaba-pai/Wan2.2-Fun-A14B-Control - https://huggingface.co/alibaba-pai/Wan2.2-Fun-5B-Control - https://huggingface.co/alibaba-pai/Wan2.2-Fun-5B-InP - https://huggingface.co/alibaba-pai/Wan2.2-Fun-A14B-InP - https://huggingface.co/alibaba-pai/Wan2.2-VACE-Fun-A14B - https://huggingface.co/Wan-AI/Wan2.2-I2V-A14B - https://huggingface.co/Wan-AI/Wan2.2-S2V-14B - https://huggingface.co/Wan-AI/Wan2.2-T2V-A14B - https://huggingface.co/Wan-AI/Wan2.2-TI2V-5B…
Open weights
apache-2.0
diffusion-single-file
LTX-2.5 is an open world model with open weights, built for local execution and fine-tuning. Its established use is generating synchronized, high-fidelity video and audio from text, image, and video inputs; applicability to emerging domains such as robotics and physical AI is developing. Full control and customization — self-host on your own infrastructure. No per-generation billing, no per-seat lock-in, no forced API dependency. Revenue is measured across the whole entity, including subsidiaries and affiliates under common control. The full, binding terms live in LICENSE. - Native multishot generation — generate connected scenes in a single pass: multiple shots that hold character…
Access requested at publisher
other
diffusion-single-file
Please check our repository or the LightX2V MiniMax-H3 examples to reproduce the results. Please check the model specifications for more details. Try the MiniMax-H3 Turbo LoRA directly in LightX2V Studio: The Studio currently uses the FL2V 8-step v1.0 768p LoRA, which provides improved video and audio generation quality with 8-step inference. Integrate MiniMax-H3 Turbo into your application through the LightX2V API
Open weights
apache-2.0
diffusers
This model card focuses on the LTX-2.3 model, which is a significant update to the LTX-2 model with improved audio and visual quality as well as enhanced prompt adherence. LTX-2 was presented in the paper LTX-2: Efficient Joint Audio-Visual Foundation Model. If you want to dive in right to the code - it is available here. LTX-2.3 is a DiT-based audio-video foundation model designed to generate synchronized video and audio within a single model. It brings together the core building blocks of modern video generation, with open weights and a focus on practical, local execution. LTX-2.3 is accessible right away via the API Playground. You can use the models - full, distilled, upscalers and any…
Open weights
other
diffusers
This model card focuses on the model associated with the LTX-Video model, codebase available here. LTX-Video is the first DiT-based video generation model capable of generating high-quality videos in real-time. It produces 30 FPS videos at a 1216×704 resolution faster than they can be watched. Trained on a large-scale dataset of diverse videos, the model generates high-resolution videos with realistic and varied content. You can use the model for purposes under the license: - 2B version 0.9: license - 2B version 0.9.1 license - 2B version 0.9.5 license - 2B version 0.9.6-dev license - 2B version 0.9.6-distilled license - 13B version 0.9.7-dev license - 13B version 0.9.7-dev-fp8 license…
Open weights
other
1.9B parameters
diffusers
This model card focuses on the LTX-2 model, as presented in the paper LTX-2: Efficient Joint Audio-Visual Foundation Model. The codebase is available here. LTX-2 is a DiT-based audio-video foundation model designed to generate synchronized video and audio within a single model. It brings together the core building blocks of modern video generation, with open weights and a focus on practical, local execution. LTX-2 is accessible right away via the following links: You can use the models - full, distilled, upscalers and any derivatives of the models - for purposes under the license. We recommend you use the built-in LTXVideo nodes that can be found in the ComfyUI Manager. For manual…
Open weights
other
18.9B parameters
diffusers
Minimax-h3Singularity is a comprehensive fine-tuned fusion model specialized in enhancing the capabilities of MiniMax-H3. Designed as a versatile multimodal video generation model, it natively supports Text-to-Video (T2V), Image-to-Video (I2V), Reference-to-Video (Ref2V), and Video-to-Video (V2V) workflows within ComfyUI. Built upon a strategic fusion of key checkpoints (including ref, fl, b25-49, etc.), this model underwent deep high-step fine-tuning. To preserve the original model's foundational strengths and broad generalization while solving artifacts introduced by high-step training, we spent 3 full days on precise model pruning and weight optimization. The result is a clean, sharp…
Open weights
apache-2.0
minimax-h3
Model · Image to video
Jay
This repository (Abiray/MiniMax-H3-GGUF) provides GGUF quantized versions and necessary component files for the MiniMax H3 model. MiniMax H3 is a general-purpose, omni-modal generative system that supports unified understanding of multimodal contexts composed of text, images, video, and audio. It can generate video with native stereo audio at resolutions up to 2K and durations of up to 15 seconds. If you are looking for a smaller model with the same great quality that fits better on consumer-tier GPUs, please check out the MiniMax-H3-Pruned-GGUF repository. The pruned architecture is compressed down to 8.9 GB – 21.6 GB, bringing MiniMax H3 execution directly to consumer hardware. This…
Open weights
other
This is the FP8 versions of the LTX-2.3 model. All information below is derived from the base model. This model card focuses on the LTX-2.3 model, which is a significant update to the LTX-2 model with improved audio and visual quality as well as enhanced prompt adherence. LTX-2 was presented in the paper LTX-2: Efficient Joint Audio-Visual Foundation Model. If you want to dive in right to the code - it is available here. LTX-2.3 is a DiT-based audio-video foundation model designed to generate synchronized video and audio within a single model. It brings together the core building blocks of modern video generation, with open weights and a focus on practical, local execution. LTX-2.3 is…
Open weights
other
diffusers
This is a GGUF quantized version of LTX-2.3. unsloth/LTX-2.3-GGUF uses Unsloth Dynamic 2.0 methodology for SOTA performance. - Important layers are upcasted to higher precision. - Uses tooling from ComfyUI-GGUF by city96. There are two sets of GGUF's published. One for the dev model and one for the distilled. The distilled model is optimized for few step generation, think 4-8 steps. dev on the other hand needs more steps at least 20, but you get better outputs. The distilled variant is useful as a drafting model or a refining model. In fact the workflow published below, uses the distilled lora on top of the dev model to refine the intial output. Download the mp4 in the repo and open it with…
Open weights
other
ggml
Model · Image to video
Joey
GGUF quantizations of the Qwen3-VL-32B vision-language text encoder used by MiniMax-H3 in ComfyUI. The H3 DiT quants are here: joeygambino/MiniMax-H3-GGUF. You need one file from each repo to run H3 — the DiT alone will not generate anything. Load these encoders with H3 Clip Loader (Any) from not the stock CLIPLoaderGGUF node. The H3 text encoder is a truncated Qwen3-VL-32B - 50 layers, no final norm, no lmhead - and its vision tower ships separately as the -mmproj-F16.gguf sidecar. Stock ComfyUI-GGUF only merges an mmproj when the encoder's architecture is qwen2vl; Qwen3-VL reports qwen3vl, so the sidecar is never merged at all, and the resulting missing vision tensors surface as a…
Open weights
gguf
LTX-2.5 is an open world model with open weights, built for local execution and fine-tuning. Its established use is generating synchronized, high-fidelity video and audio from text, image, and video inputs; applicability to emerging domains such as robotics and physical AI is developing. Full control and customization — self-host on your own infrastructure. No per-generation billing, no per-seat lock-in, no forced API dependency. Revenue is measured across the whole entity, including subsidiaries and affiliates under common control. The full, binding terms live in LICENSE. - Native multishot generation — generate connected scenes in a single pass: multiple shots that hold character…
Open weights
other
diffusion-single-file
The license of the quantized files follows the license of the original model: These files are converted using https://github.com/leejet/stable-diffusion.cpp This model can be used with stable-diffusion.cpp. For setup instructions and usage details, please refer to: To use this model in ComfyUI, first install the following custom node: An example ComfyUI workflow is available here: src="https://huggingface.co/leejet/MiniMax-H3-GGUF/resolve/main/assets/example.mp4" controls muted
Open weights
This repository (molbal/MiniMax-H3-GGUF) provides GGUF quantized versions and necessary component files for the MiniMax H3 model. MiniMax H3 is a general-purpose, omni-modal generative system that supports unified understanding of multimodal contexts composed of text, images, video, and audio. It can generate video with native stereo audio at resolutions up to 2K and durations of up to 15 seconds. This repository includes quantized versions of both the FL2VA (First-and-last-frame mode) and Ref2VA (Omni-reference mode) base models. The FL2VA builds are pruned to FP8 first and then quantized to GGUF. minimaxh3fl2vaprunedfp8Q40.gguf (11.4 GB) minimaxh3fl2vaprunedfp8Q80.gguf (20.2 GB)…
Open weights
other
LTX2.5 temp v3 has 10eros audio while v2 is sulphur audio, hopefully the 85 can fix it if you have audio problems,LTX2.5DEV10Eros15r512.safetensors seems most stable LTX2.3DISTILLEDBAKEDLTXSULPHURSTYLEIS10Erosr256.safetensors this seems to be the best for stacking lora i'm just guessing the strength and not 100% sure which reason is best 10 Eros v1.4 Changelog: Built off 1.3 and bringing back explicit prompting and motion hopefully without any kind of anatomy redraw or negative tendency. Still requires intense prompt refinement. This version is set up to be trained on to fix it into a real base, it doesn't depict anatomy well but it also isn't confused by it which is priority for the first…
Open weights
other
diffusers
Model · Image to video
Joey
GGUF quantizations of MiniMax-H3's 33B video+audio DiTs, built from MiniMax's pruned checkpoints. Same model, same quant tiers as the original-form repo, about 40% smaller - and a Q80 that fits a 24 GB card. fl2va = text/first-last-frame to video+audio (T2V and I2V). ref2va = reference-conditioned generation (identity from up to 9 images, 3 videos with soundtracks, 3 voice clips). Runs in ComfyUI with ComfyUI-GGUF plus a one-line architecture patch (node pack: ComfyUI-H3-Multishot; workflows: On anything older these will not load at all, because the shape of the modulation weights changed and older builds do not know how to read them. That is the only catch - everything else is a drop-in…
Open weights
other
minimax-h3
LTX-2.5 is an open world model with open weights, built for local execution and fine-tuning. Its established use is generating synchronized, high-fidelity video and audio from text, image, and video inputs; applicability to emerging domains such as robotics and physical AI is developing. Full control and customization — self-host on your own infrastructure. No per-generation billing, no per-seat lock-in, no forced API dependency. Revenue is measured across the whole entity, including subsidiaries and affiliates under common control. The full, binding terms live in LICENSE. - Native multishot generation — generate connected scenes in a single pass: multiple shots that hold character…
Open weights
other
diffusion-single-file
This GGUF file is a direct conversion of Lightricks/LTX-2.3 is a quantized model, all original licensing terms and usage restrictions remain in effect.
Open weights
other
gguf
Rank-32 LoRAs for MiniMax-H3 Ref2VA. Give it a photo, get a smooth clockwise 360-degree camera orbit around the frozen subject (ORB360CW), or, with the CardSpin files, turn the photo in space like a thin physical card (ORB360CARDSPIN). Both CardSpin files can still orbit with the orbit prompt, but they are noticeably softer; see Powered by MiniMax H3. poster="https://huggingface.co/MATLOWAI/MiniMax-H3-ORB360-CardSpin/resolve/main/assets/whitecatcardspinv2step50poster.jpg"> minimaxh3orb360cardspinv2step50.safetensors, one reference image (examples/whitecat.png), the prompt in prompts/cardspincaption.txt, 1024 x 768, 124 frames, 20 steps, seed 20260926.…
Open weights
other
minimax-h3
Turn two ordinary photos into the viral "Hotel Lobby" / Migos AI video with your faces. Same orange booth, same moves, same beat. No prompts, no source clip, no editing. Renders in about 3 minutes, and your first 5-second preview is free. Every clip below started as two separate photos. Sound on. Two men in their seventies with the full Quavo and Takeoff choreography. Two photos taken years apart, one performance. Office headshots in, rap duo out. More examples (couples, best friends, cats, the same person twice) on the Migos AI Video Generator page. "Migos AI", "Hotel Lobby AI" and "AI rap duo" all name the same trend: AI videos that recreate Quavo and Takeoff's 2022 performance of Hotel…
Open weights
other
other
Optimized ComfyUI workflows for MiniMax-H3, featuring INT8 inference, native Block Sparse Attention, Spectrum, Lightx2v LoRA, Turbo-LoRA, native Extend, Fun ControlNet Union V2V control, Face Refine, TTS, Latent Upscaler, Bridge, and Group Bypass. Comfy-Org/MiniMax-H3 https://huggingface.co/Comfy-Org/MiniMax-H3 The workflows use the native Block Sparse Attention node for optimized MiniMax-H3 attention processing. ComfyUI-Spectrum-MiniMax-H3 https://github.com/xmarre/ComfyUI-Spectrum-MiniMax-H3 Optimized sampling and MiniMax-H3 inference acceleration. lightx2v/Minimax-h3-Turbo https://huggingface.co/lightx2v/Minimax-h3-Turbo Acceleration LoRA for reduced sampling steps.…
Open weights
minimax-h3
Private research checkpoints — Cosmos-Predict2.5-2B robot/multiview-agibot (frameinit) post-trained on RoboCasa CoffeePressButton 3-view renders (55 human demos, 704² → trained at 480², 81 frames/view @20 fps). 2026-10. iterXXXXXX/modelemabf16.pt — EMA weights (bf16) for inference with experiment 2brobocasacoffeeagibotframeinit (code/ holds the dataset/experiment files and mvinfer.py needed to reproduce; see code/HANDOFF.md). Research only (world-model video generation for IDM labelling / policy training). Round 1 (iter004000/, identity extrinsics) showed view collapse at t≥20 frames (3 views converging to one), seed-dependent. Round 2 restarts from the agibot base with real MuJoCo camera…
Access requested at publisher
other
Community GGUF conversion of GD-ML/DreamX-Creator for the native CPU runtime in rust-model-inference. DreamX-Creator generates synchronized video and audio from a first frame and a text prompt. These files also contain the optional 2x video refiner components. Both files are required and must be used together. The runtime verifies their shared pair ID before inference. Build rust-model-inference, then run: Add --dry-run to validate the model pair and inspect the memory estimate without running generation. Full inference produces base video/audio/mux outputs and, with --refine, refined video/mux outputs. The pair is exported from the original DreamX-Creator checkpoint with the repository's…
Open weights
apache-2.0
Model · Image to video
Song
Looped Diffusion Transformer (Chng et al., 2026, arXiv:2609.40305) retrofit of Jinyan0924/flux3-agibot-videoft (itself a full fine-tune of black-forest-labs/flux-3-action-base): the 33 joint blocks are split 12 / 9 / 12, the middle 9 are run N = 4 times with head-wise gated attention, and every exit is trained with deep supervision. Task: k past head-camera frames + caption → 33 future frames at 15 fps, 384×512. Training in progress; EMA checkpoints are uploaded as they are produced (emastep-.safetensors), and full checkpoints of the compared runs (2500 and 5000 steps) will follow. Code, configs and results: https://github.com/naomili0924/looped-dit-video/tree/main/agibot-flux3…
Open weights
other
(hiddensize=[768, 2048], depth=[28, 6]) over a 64-channel geometry-native latent, at 672×378 and V=81, with metric Plücker camera conditioning and Qwen3-0.6B text. This repo is self-contained for GAE-64 inference: codec, per-channel latent statistics, DA3 feature-level stats, and the flow transformer. The frozen DA3-GIANT encoder / DPT head comes from Or, after the files are in ckpts/: Image-to-video uses 50 Euler steps and CFG 2 with one reference view. - codec/ — safetensors + config.json - gae64.pt — drop-in checkpoint for gae.loadcodec - latentstatsgae64.pt — per-channel latent mean/std - da3statsgiant5ds.tar — DA3 feature-level normalization stats - transformer/ — flow safetensors +…
Open weights
pytorch
We introduce LIFT, a unified image-to-video generation framework that complements camera control with Layout-In-FuTure control, enabling users to specify what should appear in a future view and where it should appear. Our models are built on Wan2.1-Fun-V1.1-1.3B-Control-Camera.
Open weights
apache-2.0
diffusers
Prism:Dynamic Sparse Attention for Native 2K Joint Video-Audio Generation Model Training Shuyuan Tu 1, Qi Tian 2, Yinming Huang 1, Yue Wu 2, Xintong Han 2, Kaihang Pan 3, Weijie Kong 2, Jiangfeng Xiong 2, Jian-Wei Zhang 2, Zuxuan Wu 1, Yu-Gang Jiang 1 [ 1 Fudan University; 2 Tencent Hunyuan; 3 Zhejiang University] Native training joint video-audio generation models at higher resolutions empowers them to learn richer visual details and sharper motion dynamics. However, full attention incurs quadratic cost and, as resolution increases, spreads attention over increasingly redundant tokens, diluting learning signals for informative content and disrupting pretrained priors. Existing sparse…
Open weights
mit
diffusers
ProAR solves reasoning tasks by autoregressively generating a sequence of visual states. It equips autoregressive video models with two complementary components: - Outcome Guidance predicts a future goal frame to guide current-chunk generation. - Transition Guidance aligns the current representation with the clean next-chunk representation during training. ProAR is built upon Wan2.2-TI2V-5B. hf download LinghuiShen/ProAR \ --local-dir checkpoints/ProAR
Open weights
apache-2.0
We are excited to introduce Wan2.2, a major upgrade to our foundational video models. With Wan2.2, we have focused on incorporating the following innovations: This repository also includes our I2V-A14B model, designed for image-to-video generation, supporting both 480P and 720P resolutions. Built with a Mixture-of-Experts (MoE) architecture, it achieves more stable video synthesis with reduced unrealistic camera movements and offers enhanced support for diverse stylized scenes. Your browser does not support the video tag. If your research or project builds upon Wan2.1 or Wan2.2, we welcome you to share it with us so we can highlight it for the broader community. - Wan2.2 Text-to-Video - [x]…
Open weights
apache-2.0
14.3B parameters
diffusers
Base transformer weights and their matching camera adapter and LoRA for the WorldCrafter inference code. Place this directory beside WorldCrafter-Fast. Base reads the following shared folders from Fast: repencoder/, textencoder/, tokenizer/, vae/, and scheduler/. Base does not need Fast's transformer or adapter folders. The relative path is configured in inferenceconfig.json. Keep the Base-specific transformer/ and adapter/ together. SHA256SUMS covers the packaged Base files; shared component hashes are recorded in Fast's package.
Open weights
14.3B parameters
diffusers
Self-contained weights for WorldCrafter Fast image-to-video, text-to-video, and the interactive image-to-video demo. Use the WorldCrafter inference code and its pinned uv environment. The two transformer branches have independent DMD LoRA and camera adapters. The RepEncoder, text encoder, tokenizer, VAE, and scheduler are included here. No WorldCrafter-Base files are needed for Fast inference. Image-to-video uses five high-noise steps and one low-noise step per chunk. Text-to-video uses four high-noise and eight low-noise steps for the first chunk, then two high-noise and four low-noise steps. Compilation is optional with --enable-compile. inferenceconfig.json resolves shared components…
Open weights
diffusers
