DEV Community

Cover image for New MiniMax H3 Image Preset improves errors in tiny details significantly
Furkan Gözükara
Furkan Gözükara

Posted on

New MiniMax H3 Image Preset improves errors in tiny details significantly

New MiniMax H3 Image Preset improves errors in tiny details significantly

Zoom In at 16 MP: A Real Ratchet Instead of a Fake Tool With the New MiniMax H3 Image Preset

Full tutorial link > https://www.youtube.com/watch?v=0EkDXouIiDA

Info

MiniMax H3 Image makes 16 megapixel pictures in only 6 steps on your own NVIDIA GPU, and it now has a second text to image preset for small objects: "MiniMax H3 Image - Text To Image 16 MP - KJNodes Empty Image - 6 Steps - 261114". It ships ready beside the standard preset in our SECourses ComfyUI installer V152 and in Amazing SwarmUI Presets v90, and in SwarmUI it is one checkbox of our MiniMaxH3Image extension. On many images it draws small objects more correctly: gem settings, watch gears, tools, bike sprockets and plated dishes. The model, 16 MP, 6 steps and MiniMax's official prompt enhancer running locally stay the same. Faces and rooms often look glossier and more plastic with it, so the standard preset stays the choice for portraits and natural skin.

Links:

You will learn which preset fits which picture, from same-caption, same-seed pairs: each comparison shows the whole 16 MP picture first and then zooms into the small part where the two presets differ. In ComfyUI you open the new preset from the Presets folder with Ctrl+O, see that the prompt box, output size, prompt enhancer and seed number match the standard preset, and find the one difference: the MiniMax H3 Prompt Weight node of KJNodes in "text with empty image" mode, which shows the model an empty black picture beside your text for the first quarter of the steps (2 of 6); the other 4 steps run on your text alone. Then you type a short idea, keep the enhancer on, click Run and look closely at the watch movement of the 16 MP result. In SwarmUI you filter the Presets tab for H3 Image, reset the parameters in Quick Tools, Direct Apply the new card and find its KJNodes Empty Image checkbox, which also works with your own settings. Finally you generate a jeweler at 16 MP and zoom into the ring.

Main topics include MiniMax H3 Image, KJNodes, ComfyUI and SwarmUI presets, text to image, small object detail, jewelry, watch and food images, the local prompt enhancer and 16 MP.

Use the chapters to jump to the comparisons, the ComfyUI steps or the SwarmUI steps.

Chapters:

  • 00:00:00 New preset fixes a jammed diamond setting at 16 MP
  • 00:00:24 Small objects compared: watch gears, ratchet, sprockets, dishes
  • 00:00:55 The trade-off: glossier, more plastic faces and rooms
  • 00:01:06 Which preset to use for which image
  • 00:01:31 ComfyUI: open the KJNodes Empty Image preset
  • 00:01:51 The one new box: KJNodes empty black picture
  • 00:02:07 Generate a watchmaker image with the new preset
  • 00:02:52 Inspect the watch gears in the 16 MP result
  • 00:03:05 SwarmUI: apply the new preset card
  • 00:03:30 The KJNodes Empty Image switch in SwarmUI
  • 00:03:40 Generate a jeweler image and zoom into the diamond
  • 00:04:28 Recap: which preset for which image

Extra Info

Method credit: the "text with empty image" mode of the MiniMax H3 Prompt Weight node in KJNodes by Kijai. We tested it against our standard preset on 40 same-caption, same-seed pairs at 16 MP and ship it as a second preset.

Music: Champion by Telecasted, Ghibli Station by The Mini Vandals, Curadora by Casa Rosa, Chat Roulette by Single Friend, Swing Haven 4 - Tulsa by Reed Mathis, Vespers by Topher Mohr and Alex Elena, On The Beach by Telecasted, TrueArtRealAffectionPart4 by Noir Et Blanc Vie.

Which preset when: the new preset when small objects carry the picture (watch movements, rings and gem settings, tools, sprockets, plated dishes); the standard preset for portraits, natural skin and interiors that should look like a photograph. For an important image, run both with the same prompt and keep the better one. The new preset is text to image only; for photo edits keep the Image Edit preset. No model yet? The MiniMax-H3 Core Bundle in the Model Downloader fetches every file; the Low VRAM Bundle needs less GPU memory.

This video is for ComfyUI and SwarmUI users who make product, jewelry, food and mechanism images with MiniMax H3 Image. New to ComfyUI or SwarmUI? Start with the full tutorials linked above; already installed, the update tutorial shows how to get the new presets.

For support, questions, feature requests or installation issues, see the pinned comment, our Patreon posts, our Discord and the video comments. Thank you for watching.


14 November 2026 Update V152 for ComfyUI

New: MiniMax H3 Image with KJNodes' empty-image method, for cleaner skin, hands and fine detail. A new Text To Image preset shows the model an empty picture during the first steps, a method from kijai's KJNodes. On many prompts it gives more natural skin, cleaner hands and sharper small objects than the standard preset, at the same 16 MP and with the same prompt enhancer.

The prompt enhancer is now repeatable, too: a fixed seed gives the same caption and the same image every time, in every MiniMax H3 Image preset.

The complete Presets folder of V152, the new preset marked NEW and the improved ones UPDATED:

Explorer-style list of the complete Presets folder of ComfyUI_V152.zip with real file names and sizes; the new MiniMax H3 Image KJNodes Empty Image preset is marked NEW; the MiniMax H3 Image Text To Image and Image Edit presets and the updated Krea 2, Phantom Wan, LTX2.3 Licon, LTX2.5 Text or Image To Video, FLUX 2 Low RAM and FLUX 3 Klein Advanced Image Edit presets are marked UPDATED

Every image in this post comes from a real run of the shipped presets with their default settings.

NEW PRESET — TEXT TO IMAGE WITH KJNODES EMPTY IMAGE

"MiniMax H3 Image - Text To Image 16 MP - KJNodes Empty Image - 6 Steps - 261114"

Cleaner skin, hands and fine detail from the same short prompt. Write your idea and press Run, exactly like the standard Text To Image preset. It uses the same models, all in the MiniMax-H3 Core Bundle:

Model downloader with the MiniMax-H3 Core Bundle open: 30 files, 115.68 GB, with callouts on the MiniMax H3 Image INT8 ConvRot model, the shared text encoder, the Qwen3.5 9B prompt enhancer, the MiniMax H3 Image VAE and the Download All Models button

Already using MiniMax H3 Image? There is nothing new to download. The MiniMax-H3 Low VRAM Bundle has the smaller W4A8 versions.

The same caption and seed, the standard preset on the left and the new one on the right, with details at their real pixels below. A watchmaker at his bench: the watch case, the gears and the loupe come out clean:

Comparison of the same caption and seed: a watchmaker with a loupe assembling a watch movement, made with the standard MiniMax H3 Image Text To Image preset and with the new KJNodes Empty Image preset, with the hands, tweezers and movement at 100 percent below; the new preset shows a clean round watch case, a natural loupe and clean gears

A chef plating a dish for a restaurant ad, with clean hands and a detailed plate:

Comparison of the same caption and seed: a bearded head chef plating a dish at the pass, made with the standard preset and with the new KJNodes Empty Image preset, with the hands and the plated dish at 100 percent below

An olive farmer for a brand campaign: natural, weathered skin instead of a crackled texture:

Comparison of the same caption and seed: close-up portrait of an elderly olive farmer in a flat cap, made with the standard preset and with the new KJNodes Empty Image preset, with the skin at 100 percent below; the new preset shows natural weathered skin

The empty picture also changes the composition, like a new seed, so keep both presets and use the one you like for each prompt.

Where to find it: open the preset, write your idea in the gallery box and press Run:

The MiniMax H3 Image KJNodes Empty Image preset in ComfyUI after a real run, with numbered callouts on the prompt gallery with a watchmaker idea, the prompt enhancer, the box with the prompt the model received, the KJNodes Empty Image node and the 4096 x 4096 watchmaker result

The box under the result shows the exact prompt the model received. This preset makes new images; to edit your photos, use the Image Edit preset.

UPDATED — MINIMAX H3 IMAGE TEXT TO IMAGE AND IMAGE EDIT

  • A fixed seed repeats your image: the prompt enhancer now writes the same caption for the same seed, so the image repeats too.
  • Tidier layout: the model loaders no longer overlap the nodes next to them.

UPDATED — KREA 2, PHANTOM WAN, LTX AND FLUX PRESETS

  • Krea 2: all three presets use the BF16 text encoder of the Krea 2 Core Bundle under its own file name. Already using Krea 2? Download the Krea 2 Core Bundle once more with SwarmUI Model Downloader V193 to add it. Raw Base also has a negative prompt box now.
  • Phantom Wan 14B Character Reference and LTX2.3 Licon MSR V2: the reference image switches open with the right on and off settings.
  • FLUX 2 Low RAM: the VRAM cleanup before sampling is optional now and off by default; switch it on if you run out of VRAM.
  • LTX2.5 Text or Image To Video and FLUX 3 Klein Advanced Image Edit: they open without missing example images.

ONE INSTALLER FOR EVERY PRESET

  • Windows_Install_Or_Update_ComfyUI.bat installs every custom node our presets use, and so do the RunPod and Massed Compute installers: no bundle menu, no separate node installer.
  • IP-Adapter and FaceID models now come from the new IP-Adapter And FaceID Bundle in our SwarmUI Model Downloader.
  • Installer fixes: a moved, renamed or copied ComfyUI folder gets a working environment again, and a failed step is reported as an error.

HOW TO UPDATE

  1. Download the latest ComfyUI_V152.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. It installs and updates every node our presets use.
  3. If you do not have it yet, download the MiniMax-H3 Core Bundle (or the Low VRAM Bundle) with our SwarmUI Model Downloader V190 or newer.
  4. Restart ComfyUI and refresh the browser tab.

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


14 November 2026 Update V193 for SwarmUI

New in SwarmUI: MiniMax H3 Image with KJNodes' empty-image method, for cleaner skin, hands and fine detail. A new preset shows the model an empty picture during the first steps, a method from kijai's KJNodes. On many prompts it gives more natural skin, cleaner hands and sharper small objects than the standard Text To Image preset, at the same 16 MP and with the same prompt enhancer. It is also a new switch, MiniMax H3 Image KJNodes Empty Image, so you can turn it on with your own settings.

The prompt enhancer is now repeatable, too: a fixed seed gives the same caption and the same image every time.

Amazing SwarmUI Presets moves to V90 with 102 presets: 1 new and 5 updated.

All 102 presets in the Presets tab, the new one marked NEW and the updated ones UPDATED:

SwarmUI Presets tab in Cards view with all 102 Amazing SwarmUI Presets V90 and their thumbnails; the new MiniMax H3 Image KJNodes Empty Image preset is marked NEW and the MiniMax H3 Image Text To Image and Image Edit presets and the three Krea 2 presets are marked UPDATED

The new and updated presets up close; the new one has its own great hornbill thumbnail:

The six new and updated preset cards in the SwarmUI Presets tab: Krea 2 Raw/Base Slow with a brown bear, Krea 2 Turbo Fast with a black panther, Krea 2 Turbo Image Edit with a tortoise, MiniMax H3 Image Text To Image 16 MP with a giant panda and Image Edit Up To 16 Images with a red-crowned crane, all marked UPDATED, and MiniMax H3 Image Text To Image 16 MP KJNodes Empty Image with a great hornbill, marked NEW, with their descriptions

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

NEW PRESET — TEXT TO IMAGE WITH KJNODES EMPTY IMAGE

"MiniMax H3 Image - Text To Image 16 MP - KJNodes Empty Image - 6 Steps - 261114"

Cleaner skin, hands and fine detail from the same short prompt. Apply it, write your idea and click Generate, exactly like the standard Text To Image preset. It uses the same models, all in the MiniMax-H3 Core Bundle:

Model downloader with the MiniMax-H3 Core Bundle open: 30 files, 115.68 GB, with callouts on the MiniMax H3 Image INT8 ConvRot model, the shared text encoder, the Qwen3.5 9B prompt enhancer, the MiniMax H3 Image VAE and the Download All Models button

Already using MiniMax H3 Image? There is nothing new to download. The MiniMax-H3 Low VRAM Bundle has the smaller W4A8 versions.

The same caption and seed, the standard preset on the left and the new one on the right, made with the same presets in our ComfyUI, with details at their real pixels below. A watchmaker at his bench: the watch case, the gears and the loupe come out clean:

Comparison of the same caption and seed: a watchmaker with a loupe assembling a watch movement, made with the standard MiniMax H3 Image Text To Image preset and with the new KJNodes Empty Image preset, with the hands, tweezers and movement at 100 percent below; the new preset shows a clean round watch case, a natural loupe and clean gears

A chef plating a dish for a restaurant ad, with clean hands and a detailed plate:

Comparison of the same caption and seed: a bearded head chef plating a dish at the pass, made with the standard preset and with the new KJNodes Empty Image preset, with the hands and the plated dish at 100 percent below

An olive farmer for a brand campaign: natural, weathered skin instead of a crackled texture:

Comparison of the same caption and seed: close-up portrait of an elderly olive farmer in a flat cap, made with the standard preset and with the new KJNodes Empty Image preset, with the skin at 100 percent below; the new preset shows natural weathered skin

The empty picture also changes the composition, like a new seed, so keep both presets and use the one you like for each prompt.

Where to find it: Presets tab → filter 261114 → apply the preset, write your idea and click Generate:

The MiniMax H3 Image KJNodes Empty Image preset in SwarmUI after a real run, with numbered callouts on the preset card in the Presets tab, the short prompt about an olive farmer, the prompt enhancer and the KJNodes Empty Image switch in the MiniMax H3 Image group, and the 3648 x 4576 olive farmer result

The switch works for text to image; when you attach images to the prompt, the preset edits them as usual.

UPDATED — MINIMAX H3 IMAGE TEXT TO IMAGE AND IMAGE EDIT

  • A fixed seed repeats your image: the prompt enhancer now writes the same caption for the same seed, so the image repeats too.
  • Both presets keep the new switch off, so they work as before.

UPDATED — KREA 2

All three Krea 2 presets now select the BF16 text encoder of the Krea 2 Core Bundle under its own file name, so they always get it instead of SwarmUI's smaller default. Already using Krea 2? Download the Krea 2 Core Bundle once more to add it.

NEW IN THE MODEL DOWNLOADER

  • IP-Adapter And FaceID Bundle (SD 1.5, SDXL, Kolors): every IP-Adapter, FaceID adapter, FaceID LoRA and image encoder in one click, for SwarmUI's IP-Adapter feature and for ComfyUI.
  • Wan 2.2 Core 4 Steps Bundle now includes RIFE 4.7 frame interpolation.
  • LTX 2.5 Upscaler bundle: the smaller INT8 text encoder that SwarmUI uses by default, plus the Gemma 4 E2B prompt enhancer.
  • Download queue: shows how many downloads have finished, failed or been cancelled.
  • SwarmUI installer: if SwarmUI cannot be installed, it says why and what to do next.

HOW TO UPDATE

  1. Download SwarmUI_Model_Downloader_v193.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.
  3. Update the ComfyUI backend: close SwarmUI again (step 2 starts it) and run Windows_Install_Or_Update_ComfyUI.bat from our ComfyUI V152.
  4. Restart SwarmUI, then run Windows_Preset_Delete_Import.bat to import Amazing_SwarmUI_Presets_v90.json. It backs up your current presets first, then replaces them all, your own presets too.
  5. If you do not have it yet, run Windows_Start_Download_Models_App.bat and download the MiniMax-H3 Core Bundle, or the MiniMax-H3 Low VRAM Bundle for less memory.

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

Top comments (0)