Home Creators Posts Import Register Favorites Logout

Content

LongCat Avatar 1.6 turns one photo and one audio file into a lip-synced talking video on your own GPU; our 3-step preset runs 2.9x faster than 8 steps.

Ships as the 3 Steps Really Fast preset for ComfyUI (installer V154) and SwarmUI (presets pack v92), and SwarmUI Model Downloader v195 gets the 11-model core bundle (almost 40 GB) in one click. A 15-second 720p clip took 4 min 24 s instead of 12 min 35 s on one RTX 5090, and one run makes a two-minute video.

Full tutorial link > https://www.youtube.com/watch?v=6LkyNG0Xq9A

Turn Your Photo and Voice Into a Talking Video on Your Own GPU With LongCat

Our new LongCat Avatar 1.6 "3 Steps Really Fast" preset turns one photo and one audio file into a lip-synced talking video 2.9 times faster than the default 8 steps, with nearly the same picture: 4 min 24 s instead of 12 min 35 s for the same 15-second 720p clip on one RTX 5090 (conditions below). We built this 3-step pipeline ourselves; it ships for ComfyUI (our installer, V154) and SwarmUI (presets pack v92), and one click in our SwarmUI Model Downloader (v195) gets every model. You also see MiniMax H3 break on 30 seconds of audio while LongCat keeps going, up to a two-minute lesson made in one run.

Links:

Info

You will learn how to open the preset in ComfyUI with Ctrl+O, upload the soundtrack (the video is as long as the audio) and the photo, keep the default scene prompt or type one short line, see what makes it fast, choose 720p, 480p or a custom size and press Run. In SwarmUI you filter the presets for really fast, reset the parameters, Direct Apply the preset, attach the chef's audio and photo and press Generate. Finally you download the LongCat Avatar core bundle (11 models, almost 40 GB) with one click.

Use the chapters to jump to install steps, setup, generation settings and advanced features.

Chapters:

  • 00:00:00 Talking video from one photo, 2.9x faster than 8 steps

  • 00:00:25 Default 8 steps vs our 3 steps on one clip

  • 00:00:40 A two-minute math lesson made in one run

  • 00:00:51 LongCat Avatar 1.6 Really Fast preset for ComfyUI and SwarmUI

  • 00:01:03 The 3-step pipeline we built ourselves

  • 00:01:29 RTX 5090 speed test: 3, 4 and 8 steps

  • 00:01:54 Face close-up: lip sync and detail across step counts

  • 00:02:06 Our MiniMax H3 lip sync preset at 832 x 1248

  • 00:02:22 30 seconds of audio: MiniMax H3 vs LongCat 3 steps

  • 00:02:48 One-minute and two-minute talking videos in a single run

  • 00:03:17 Why any length works: segments that continue each other

  • 00:03:29 ComfyUI: open the 3 Steps Really Fast preset with Ctrl+O

  • 00:04:01 Upload the audio and photo, add a scene prompt

  • 00:04:47 What makes it fast: 3 steps and INT8 attention

  • 00:05:00 Size modes: 480p, 720p or a custom size

  • 00:05:41 Press Run and watch the finished barista video

  • 00:06:08 SwarmUI: apply the same preset with one click

  • 00:06:37 Add the chef's audio and photo, then Generate

  • 00:07:33 One-click download: the LongCat Avatar core bundle

  • 00:07:58 Get the ComfyUI installer and presets from the post

  • 00:08:14 Recap: any length, 2.9 times faster with 3 steps

More Info

Music: Captain by Topher Mohr and Alex Elena, Vibe Check by Blue Deer Studio, Tomorrow's Light by Telecasted, Candy Apple Blue by Surf Ninja 3, Ponte de Abril by Steve Adams, In it to Win it by Dan "Lebo" Lebowitz, Tone Seeker, Say Yeah by Topher Mohr and Alex Elena, Night Run Away by An Jone.

Tested November 19, 2026 on one RTX 5090 32 GB (ComfyUI V154 with --reserve-vram 3, presets as shipped, same photo, 15.36 s audio and seed, 800 x 1152, one job at a time): 8 steps 12 min 35 s, 4 steps 6 min 33 s, 3 steps 4 min 24 s with the models loaded (2.9 times faster). Our 3-step times apply that day's LongCat node update at 8.2 percent faster, its smaller measured gain. AI demo clips are shown upscaled 4x with RelayVSR (SECourses Premium Upscaler); times are generation times. SSIM against 8 steps: 0.87 (3 steps), 0.89 (4 steps). 3 steps: 31.5 s of audio in 9 min 23 s, 60 s in 17 min 9 s, 119.6 s in 34 min 24 s. MiniMax H3 Lip Synch Turbo 4 Steps at 832 x 1248: 15 s in 4 min 27 s; the 31.5 s chef audio ran 12 min 20 s and returned grey noise. The barista run in the video took about five minutes with the screen recording running.

Our pipeline: one premerged model with an exact in-place switch between the DMD and OPSD merges (no LoRA work at run time), three of the official eight distilled steps, INT8 attention and a last segment only as long as the audio. 3.7-second segments continue each other with your photo as the identity anchor, and the saved video keeps your audio.

This video is for creators, marketers, teachers and small businesses who want adverts, lessons and presenter clips from a photo and a voice on their own NVIDIA GPU. New to ComfyUI or SwarmUI? Start with the tutorials above. Thank you for watching.

18 November 2026 Update V154 - ComfyUI

New: LongCat-Video-Avatar 1.6 Really Fast. A new preset turns one photo and any audio into a lip-synced 720p talking or singing video of any length, with 3 of the official 8 distilled steps. And one new premerged model now runs every LongCat preset.

The complete Presets folder of V154, the new preset marked NEW and the updated one UPDATED:

Every example in this post is a real LongCat-Video-Avatar 1.6 result, nothing retouched.

NEW PRESET — LONGCAT AVATAR 1.6, 720P, 3 STEPS REALLY FAST

"LongCat Avatar 1.6 - Image + Audio - 720p 3 Steps Really Fast - Premerged DMD + OPSD - 261118"

Your photo and your voice become a 720p talking or singing video of any length, with 3 of the official 8 steps. The last segment is made only as long as your audio needs, and the faster INT8 attention is already on. All models come in the LongCat-Video-Avatar 1.6 Core Bundle, first in our downloader:

The bundle's new default model is our premerged INT8 ConvRot: both distillation LoRAs are merged in, so no LoRA work happens while sampling, and one model runs every LongCat preset. The plain INT8 ConvRot and BF16 models stay available in the downloader.

One photo and a 15-second voice recording, made with the new preset at its default 720p:

The same man, glasses, T-shirt and room from the first second to the last, while his face, mouth and hands move with the words.

The same moments of the official 8 steps and the new 3-step preset, made from the same photo, speech, seed and size:

In our test with this 15-second 720p clip, the 3-step preset made the video about 2.9 times faster than the official 8 steps, with the same person, glasses, room and lip sync.

All our extras are kept: the prompt box with + Prompt and folder batch, soundtrack load and trim, the optional photo, the optional LoRA stack, live preview, the mouth enhancement switch, and a note beside every setting:

Upload your audio and photo; everything else is already set.

Where to find it: open the preset, upload your audio and photo, and press Run:

The finished video plays in the preview with your audio and is saved in ComfyUI's output folder under video/LongCat_Avatar.

UPDATED PRESET — LONGCAT AVATAR 1.6, 8 STEPS

"LongCat Avatar 1.6 - Audio + Image or Text To Long Lip Synch Video - 8 Steps - 261117"

The official 8-step recipe now runs on the same premerged model. Same look: on the same photo, audio and seed it makes the same video as before, and you keep only one LongCat model on disk.

HOW TO UPDATE

  1. Download the latest ComfyUI_V154.zip from this post and extract it over your installer folder.

  2. Close ComfyUI, and SwarmUI too if it uses this ComfyUI, then run Windows_Install_Or_Update_ComfyUI.bat.

  3. Download the LongCat-Video-Avatar 1.6 Core Bundle with our SwarmUI Model Downloader V195.

  4. Restart ComfyUI and refresh the browser tab.

SwarmUI Installer and Unified Massive Models Downloader: https://www.patreon.com/posts/114517862

18 November 2026 Update V195

New in SwarmUI: LongCat-Video-Avatar 1.6 Really Fast. A new preset turns one photo and any audio into a lip-synced 720p talking or singing video of any length, with 3 of the official 8 distilled steps. One new premerged model now runs every LongCat preset, and the downloader shows every bundle in one tab.

Amazing SwarmUI Presets moves to V92 with 105 presets: 1 new and 2 updated.

All 105 presets in the Presets tab, the new one marked NEW and the updated ones UPDATED:

The three LongCat presets up close, each with its own thumbnail:

Windows_Preset_Delete_Import.bat brings all of them into your SwarmUI; the steps are in How to update at the end.

NEW PRESET — LONGCAT AVATAR 1.6, 720P, 3 STEPS REALLY FAST

"LongCat Avatar 16 - Audio + Image To Lip Sync Video - 720p - 3 Steps Really Fast - 261118"

Your photo and your voice become a 720p talking or singing video of any length, with 3 of the official 8 steps. Apply the preset, attach the speech or song to the prompt, set the person's photo as Init Image and click Generate. All models come in the LongCat-Video-Avatar 1.6 Core Bundle, first in the new ComfyUI & SwarmUI Bundles tab of the downloader:

The bundle's new default model is our premerged INT8 ConvRot: both distillation LoRAs are merged in, so no LoRA work happens while sampling, and one model runs every LongCat preset. The plain INT8 ConvRot and BF16 models stay available in the downloader.

One photo and a 15-second voice recording, made with the same 3-step recipe at 720p in our ComfyUI:

The same man, glasses, T-shirt and room from the first second to the last, while his face, mouth and hands move with the words.

The same moments of the official 8 steps and the new 3-step preset, made from the same photo, speech, seed and size:

In our test with this 15-second 720p clip, the 3-step preset made the video about 2.9 times faster than the official 8 steps, with the same person, glasses, room and lip sync.

New in the LongCat Avatar group: Distilled Step Count (8, 4 or 3), Shorten Last Segment, which makes the last segment only as long as your audio needs, and int8 attention. The 3-step preset sets them for you. Prompt List, Mouth Enhancement and your SwarmUI LoRAs work as before.

Where to find it: Presets tab → filter LongCat → apply the 3-step preset, attach your audio, set your photo as Init Image and click Generate. Every finished video waits in History:

Reuse Parameters on a finished video loads its settings again for your next one.

UPDATED PRESETS — LONGCAT AVATAR 1.6, 8 STEPS

"LongCat Avatar 16 - Audio + Image To Lip Sync Video - 480p - 8 Steps - 261117"

"LongCat Avatar 16 - Audio + Image To Lip Sync Video - 720p - 8 Steps - 261117"

The official 8-step recipe now runs on the same premerged model. Same look: on the same photo, audio and seed it makes the same video as before, and you keep only one LongCat model on disk.

NEW IN THE MODEL DOWNLOADER

  • One 📦 ComfyUI & SwarmUI Bundles tab with every bundle: the SwarmUI bundles first, the LongCat-Video-Avatar 1.6 Core Bundle on top.

  • LongCat Video Avatar 1.6 Premerged DMD OPSD INT8 ConvRot in the Core Bundle and under Video Generation Models → LongCat-Video-Avatar 1.6 Models.

HOW TO UPDATE

  1. Download SwarmUI_Model_Downloader_v195.zip from this post and extract it over your downloader folder.

  2. Close SwarmUI and run Windows_Install_or_Update_SwarmUI.bat. It updates SwarmUI and our extensions, with the new LongCat Avatar options.

  3. Update the ComfyUI backend: close SwarmUI again (step 2 starts it) and run Windows_Install_Or_Update_ComfyUI.bat from our ComfyUI V154.

  4. Restart SwarmUI, then run Windows_Preset_Delete_Import.bat to import Amazing_SwarmUI_Presets_v92.json. It backs up your current presets first, then replaces them all, your own presets too.

  5. Run Windows_Start_Download_Models_App.bat and download the LongCat-Video-Avatar 1.6 Core Bundle.

ComfyUI Installer and Presets: https://www.patreon.com/posts/105023709

Related posts

All posts in this topic: AI Video Generation and Animation Apps · New to SECourses? Start Here: Best and Most Popular SECourses Posts

Files

Turn Your Photo and Voice Into a Talking Video on Your Own GPU With LongCat

Our new LongCat Avatar 1.6 "3 Steps Really Fast" preset turns one photo and one audio file into a lip-synced talking video 2.9 times faster than the default 8 steps, with nearly the same picture: 4 min 24 s instead of 12 min 35 s for the same 15-second 720p clip on one RTX 5090 (conditions below). We built this 3-step pipeline ourselves; it ships for ComfyUI (our installer, V154) and SwarmUI (presets pack v92), and one click in our SwarmUI Model Downloader (v195) gets every model. You also see MiniMax H3 break on 30 seconds of audio while LongCat keeps going, up to a two-minute lesson made in one run. Links: Download: ComfyUI installer zip (V154) with the preset (our ComfyUI post): [ https://www.patreon.com/posts/105023709 ] SwarmUI version: presets pack v92 and Model Downloader v195 (our SwarmUI post): [ https://www.patreon.com/posts/114517862 ] SECourses Premium Upscaler app (RelayVSR): [ https://www.patreon.com/posts/secourses-pro-150202809 ] AI Avatar Model Showdown (MiniMax H3, LongCat and more): [ https://youtu.be/aOM_YEiskLg ] ComfyUI installation tutorial: [ https://youtu.be/nxsCBTcCVp4 ] SwarmUI full tutorial: [ https://youtu.be/0HmIeeI1iBo ] Update ComfyUI, SwarmUI and presets: [ https://www.youtube.com/watch?v=sdD3--WbHZk ] Windows requirements tutorial: [ https://youtu.be/ptuLL0C72Ms ] SECourses Discord: [ https://discord.com/invite/software-engineering-courses-secourses-772774097734074388 ] You will learn how to open the preset in ComfyUI with Ctrl+O, upload the soundtrack (the video is as long as the audio) and the photo, keep the default scene prompt or type one short line, see what makes it fast, choose 720p, 480p or a custom size and press Run. In SwarmUI you filter the presets for really fast, reset the parameters, Direct Apply the preset, attach the chef's audio and photo and press Generate. Finally you download the LongCat Avatar core bundle (11 models, almost 40 GB) with one click. Use the chapters to jump to install steps, setup, generation settings and advanced features. Chapters: 0:00 Talking video from one photo, 2.9x faster than 8 steps 0:25 Default 8 steps vs our 3 steps on one clip 0:40 A two-minute math lesson made in one run 0:51 LongCat Avatar 1.6 Really Fast preset for ComfyUI and SwarmUI 1:03 The 3-step pipeline we built ourselves 1:29 RTX 5090 speed test: 3, 4 and 8 steps 1:54 Face close-up: lip sync and detail across step counts 2:06 Our MiniMax H3 lip sync preset at 832 x 1248 2:22 30 seconds of audio: MiniMax H3 vs LongCat 3 steps 2:48 One-minute and two-minute talking videos in a single run 3:17 Why any length works: segments that continue each other 3:29 ComfyUI: open the 3 Steps Really Fast preset with Ctrl+O 4:01 Upload the audio and photo, add a scene prompt 4:47 What makes it fast: 3 steps and INT8 attention 5:00 Size modes: 480p, 720p or a custom size 5:41 Press Run and watch the finished barista video 6:08 SwarmUI: apply the same preset with one click 6:37 Add the chef's audio and photo, then Generate 7:33 One-click download: the LongCat Avatar core bundle 7:58 Get the ComfyUI installer and presets from the post 8:14 Recap: any length, 2.9 times faster with 3 steps Music: Captain by Topher Mohr and Alex Elena, Vibe Check by Blue Deer Studio, Tomorrow's Light by Telecasted, Candy Apple Blue by Surf Ninja 3, Ponte de Abril by Steve Adams, In it to Win it by Dan "Lebo" Lebowitz, Tone Seeker, Say Yeah by Topher Mohr and Alex Elena, Night Run Away by An Jone. Tested November 19, 2026 on one RTX 5090 32 GB (ComfyUI V154 with --reserve-vram 3, presets as shipped, same photo, 15.36 s audio and seed, 800 x 1152, one job at a time): 8 steps 12 min 35 s, 4 steps 6 min 33 s, 3 steps 4 min 24 s with the models loaded (2.9 times faster). Our 3-step times apply that day's LongCat node update at 8.2 percent faster, its smaller measured gain. AI demo clips are shown upscaled 4x with RelayVSR (SECourses Premium Upscaler); times are generation times. SSIM against 8 steps: 0.87 (3 steps), 0.89 (4 steps). 3 steps: 31.5 s of audio in 9 min 23 s, 60 s in 17 min 9 s, 119.6 s in 34 min 24 s. MiniMax H3 Lip Synch Turbo 4 Steps at 832 x 1248: 15 s in 4 min 27 s; the 31.5 s chef audio ran 12 min 20 s and returned grey noise. The barista run in the video took about five minutes with the screen recording running. Our pipeline: one premerged model with an exact in-place switch between the DMD and OPSD merges (no LoRA work at run time), three of the official eight distilled steps, INT8 attention and a last segment only as long as the audio. 3.7-second segments continue each other with your photo as the identity anchor, and the saved video keeps your audio. This video is for creators, marketers, teachers and small businesses who want adverts, lessons and presenter clips from a photo and a voice on their own NVIDIA GPU. New to ComfyUI or SwarmUI? Start with the tutorials above. Thank you for watching.

Comments

No comments found for this post.