Stable Diffusion Tutorials

@SD_Tutorial • 12,866 subscribers

👉 Ai models local installation 👉 Comfy Workflows 👉 Tutorials (Image Gen, Video gen) FOLLOW WEBSITE 👇👇

Shorts

218,676 views

251,462 views

19,573 views

18,605 views

38,333 views

23,088 views

17,187 views

29,096 views

14,369 views

13,197 views

21,628 views

12,310 views

11,023 views

Videos

sweetdream.ai

Watch Anya Live

Anya is streaming live right now! Join her private show and enjoy exclusive content.

Private Show

Join now for exclusive access

Free preview available • Premium content

399,978 views • 22 days ago

364,589 views • 2 months ago

16,965 views • 3 days ago

88,284 views • 1 month ago

78,231 views • 1 month ago

19,869 views • 11 days ago

102,477 views • 2 months ago

14,915 views • 10 days ago

48,002 views • 1 month ago

60,850 views • 2 months ago

18,655 views • 19 days ago

48,939 views • 2 months ago

18,279 views • 1 month ago

18,877 views • 1 month ago

23,115 views • 2 months ago

37,906 views • 4 months ago

15,588 views • 2 months ago

10,221 views • 1 month ago

12,737 views • 1 month ago

25,968 views • 5 months ago

Live Cam

Stable Diffusion Tutorials

Shorts

LTX 2.5 😀 Restore IC-LoRA it takes archive video (low-resolution web rips, low-bitrate broadcast transfers, tape, sepia or black-and-white film scans) as its reference and renders the same shot as a clean, colour

Apple's T800 🙄 Pro Max getting out of hand !

MiniMax-H3 🙂X2-Detail-VAE in ComfyUI contains a 2-in-1 MiniMax H3 X2 VAE, an optional ComfyUI detail-enhancement node -Mode 1 — 2X VAE Decode -Mode 2 — Reference Detail Enhancement custom node+model+workflow:👇

LTX-2.5 IC-LoRA Alpha Gen takes an ordinary RGB video and generates a matching alpha matte for it, with no green screen, masks, or prompt required. It handles hard subjects as well as soft and semi-transparent elements such as hair, fur, smoke, fire etc

MiniMax H3 💋DaSiWa (N$FW) -REF2VA + Optimized FL2VA mixture -Better Sound -Better Voices -Enhanced Quality and Reasoning -Better understanding of N$FW 👇

Minimax H3😃TaoMate LORA for Comfy 3Step LORA👇 Use this with TaoMate-H3 👇

Minimax H3 🔄 🎥 VFX Camera control Visual camera planner for MiniMax H3 inside ComfyUI. You drag the camera around a 3D sphere, place keyframes on a timeline, and the node compiles that trajectory into prompts that H3 understands. Workflow on civitAI:👇

MiniMax H3 ⚡️ Spectrum acceleration in ComfyUI — 34% lower Euler sampling time, 30% lower RES time Tested at approximately 0.5 MP, 8 seconds and 20 steps on an RTX PRO 6000 under WSL, using High VRAM mode 👇

Minimax H3😀Prompt Builder for ComfyUI -Guided prompt writing and reference-media handling -For the open-weight MiniMax H3 video model in ComfyUI. -Supports named sections, shot timings, speaker IDs, and tags pointing at your reference media 👇

Minimax H3 🤯 Amazing Prompt Heck -Inject your ref img -Circle on the reference image where you want your scene to be and get your desired result full detail in comment👇🧐

Updated: 😃LTX 2.3 Video Edit Lora📽️✂️ -Helpful for video editing, background/subject removal, motion transfer etc. -The new updated version improves on consistent results and prompt adherence Model + Workflows:👇

Character Swap with 😃Minimax H3 details in [comment]

LTX 2.3 🧐 IC-LoRA CrossView Warp 0.9 -merges a original video ref. and a camera offset (azimuth / elevation / distance), and it generates the same scene from that new viewpoint -fine tuned with 18k steps 👇

Videos

Watch Anya Live

MiniMax H3 😉ref2vid + Blender camera motion in ComfyUI workflow + details in comment👇

Minimax H3's gen 😳on RTX 6000 (1080p) T-800's in its prompt era

New Minimax H3 🙂 Character Swap LoRA V1.5 where V1 loses the reference identity or returns to the source appearance, while V1.5 retains more of the character.

Minimax H3 😀 -Advanced Prompting style with micro-expressions + emotions using tags and context for natural detailed output full details in comment

MiniMax H3 😶‍🌫️ Sharpness LORA This supports ref2VA/fl2va, sharpens a source video while keeping it photorealistic. It conditions on the source through guide latents rather than through the model's native reference-video node. 👇

MiniMax H3 🧐 Person Remover LoRA tracks the person with SAM 3.1, fills the mask green, and generates the replacement background in overlapping windows. Supply the original video and a clean version of its first frame.

MiniMax H3 🧐 Comfy Tested -Minimum requirements for 480p video on this model is a 3060 with 12GB vram, 32GB of system ram and a good nvme SSD. -generating a 5 second (124 frames) 480p (864x480) video on this system and it took a bit less than 9 minutes end to end (20 steps)

Minimax H3 😃 VEDA Sparse Attention is now in ComfyUI -Faster video Gen -A Predictor computes the relevant subset instead of the full attention map -supports with T2VA, FL2VA and R2VA Custom node: Predictor: details in comment👇

Viggle-Animate Minimax H3 😃 Fined tuned replaces video characters using just a single repainted frame. It handles the absolute hardest motions—jumps, full kicks, whipping heads without smearing through the frames 👇

Will Smith x Spaghetti remix 🤪 MiniMax-H3 on RTX 3060 6GB &amp; 64GB RAM (A 5s 480p T2V clip took around 4–5 mins using EasyCache + SageAttention)

ComfyUI-Continuity One node for AI video in ComfyUI. Write a prompt, attach media with @, press Render. Support- H3, LTX-2.5, qwen img2.1, Krea2, ideogram4 -Character Tab -RefMod save -Seg &amp; Global Seg Control -Select Continuity &amp; Seam Optimization 👇

LTX 2.5 🔥🔥🔥 Opens source model on the way.... Community is saying .... Its not a Minimax H3 rival model but a better upgrade from earlier 2.3 variant

😉Minimax H3 + Realism lora + better human motion lora + face detailer + cinematic lora details in comment 👇

MiniMax H3 🧐 offline prompt composer for video generation in ComfyUI. -for T2VA, I2VA, FL2VA, L2VA, and Ref2VA. -Reuse characters, environments, voices, and keyframes across scenes. -Custom Camera Builder, visual path planners, Advanced Editing github:

MiniMax-H3 😃 Multishot Workflow Chain multiple MiniMax-H3 shots into one continuous video with audio - in one node. Make 30 sec long video 👇

Nvidia's PID (Pixel diffusion Decoder) It decodes latent representations into high-resolution images, replacing the decode–then–super-resolve cascade while achieving lower latency and higher visual quality page: HF repo:

LTX 2.5 vs Minimax H3

Minimax H3 Viggle animate Character swap

Minimax H3 🤨 Known Characters Dataset List of Celebrity names, character titles, and pop-culture entities the H3 model already perfectly memorized (good/bad/onthefence ) &amp; which ones it will fail to render accurately without manual training 👇

LTX 2.3 v1.1 on 8GB Tested on rtx 4060ti 16gb vram and 64gb ram workflow:

Will Smith x Spaghetti remix 🤪 MiniMax-H3 on RTX 3060 6GB & 64GB RAM (A 5s 480p T2V clip took around 4–5 mins using EasyCache + SageAttention)

ComfyUI-Continuity One node for AI video in ComfyUI. Write a prompt, attach media with @, press Render. Support- H3, LTX-2.5, qwen img2.1, Krea2, ideogram4 -Character Tab -RefMod save -Seg & Global Seg Control -Select Continuity & Seam Optimization 👇

Minimax H3 🤨 Known Characters Dataset List of Celebrity names, character titles, and pop-culture entities the H3 model already perfectly memorized (good/bad/onthefence ) & which ones it will fail to render accurately without manual training 👇