Book a demo Start for free

Compare model versions

Model: Plaguekind/Minimax-H3

Earlier snapshot Snapshot #498 Repository revision 4b5b724 (opens in a new tab) Recorded at 21 Sep 2026, 06:48 UTC
Later snapshot Snapshot #534 Repository revision f1e1089 (opens in a new tab) Recorded at 22 Sep 2026, 06:39 UTC

Summary

Licence

Unchanged

No change

MIT MIT

PinDiff reports the declared licence and licence files. It does not interpret licence terms.

Evidence · Observed by PinDiff

Earlier
Snapshot #498 · revision 4b5b724 · Recorded at 21 Sep 2026, 06:48 UTC
Later
Snapshot #534 · revision f1e1089 · Recorded at 22 Sep 2026, 06:39 UTC
Source
Licence tag declared in the model card metadata, observed via the Hugging Face API; licence file hashes from the repository file manifest.

Weights

Unchanged

No change

Evidence · Observed by PinDiff

Earlier
Snapshot #498 · revision 4b5b724 · Recorded at 21 Sep 2026, 06:48 UTC
Later
Snapshot #534 · revision f1e1089 · Recorded at 22 Sep 2026, 06:39 UTC
Source
Weight file hashes and sizes from the repository file manifest (Git LFS SHA256 where available). Weight files are not downloaded.

Remote code / configuration

Unchanged

No change

Remote code No No

Evidence · Observed by PinDiff

Earlier
Snapshot #498 · revision 4b5b724 · Recorded at 21 Sep 2026, 06:48 UTC
Later
Snapshot #534 · revision f1e1089 · Recorded at 22 Sep 2026, 06:39 UTC
Source
Stored config.json and generation_config.json at each revision (trust_remote_code / auto_map), and Python file hashes from the repository file manifest.

Access / gating

Unchanged

No change

Public Public

Evidence · Observed by PinDiff

Earlier
Snapshot #498 · revision 4b5b724 · Recorded at 21 Sep 2026, 06:48 UTC
Later
Snapshot #534 · revision f1e1089 · Recorded at 22 Sep 2026, 06:39 UTC
Source
Gated and private flags observed via the Hugging Face API.

Model card

Changed

Model card modified (5 sections)

V11 Notes Added +9 −0

Diff

Added: + - UPDATE NODEPACK TO V1.5.6 AND COMFYUI TO NEWEST VERSION, and requirements.
Added: + - Added Refmod Support, create in subgraph.
Added: + - Got rid of fantastic builder kept only luisachaotica.
Added: + - Fixed layout issues and removed unwanted dependencies.
Added: + - Dropped turbo, it was ruining video and audio. default WF settings will get you 4 skips for "9" steps.
Added: + - Added my H3 Combo node that does all modes.
Added: + - Overall should be better prompt adherence, faster, and less overhead with reduced peak vram.
Added: + - Use reference map to know what the LM is calling everything.
Added: + - If you don't like no turbo, just use your favorite settings from before. Shift node in Sampler subgraph for people who want to keep using turbo.

Before

Not present in the earlier snapshot.

After

- UPDATE NODEPACK TO V1.5.6 AND COMFYUI TO NEWEST VERSION, and requirements.
- Added Refmod Support, create in subgraph.
- Got rid of fantastic builder kept only luisachaotica.
- Fixed layout issues and removed unwanted dependencies.
- Dropped turbo, it was ruining video and audio. default WF settings will get you 4 skips for "9" steps.
- Added my H3 Combo node that does all modes.
- Overall should be better prompt adherence, faster, and less overhead with reduced peak vram.
- Use reference map to know what the LM is calling everything.
- If you don't like no turbo, just use your favorite settings from before. Shift node in Sampler subgraph for people who want to keep using turbo.
Requirements Changed +2 −9

Diff

  - [ComfyUI-PlagueKind-Nodes](https://github.com/PlagueKind/Comfyui-PlagueKind-Nodes)
Removed: -
  - [ComfyUI-KJNodes](https://github.com/kijai/ComfyUI-KJNodes)
Removed: -
  - [Rgthree-comfy](https://github.com/rgthree/rgthree-comfy)
Removed: -
  - [Nvidia_RTX_Nodes_ComfyUI](https://github.com/Comfy-Org/Nvidia_RTX_Nodes_ComfyUI)
Removed: -
  - [ComfyUI-VideoHelperSuite](https://github.com/Kosinkadink/ComfyUI-VideoHelperSuite)
Removed: -
  - [ComfyUI-H3-Motion-Context-MultiRef
Removed: -
  ](https://github.com/seitanism/ComfyUI-H3-Motion-Context-MultiRef)
Removed: -
Removed: - - [ComfyUI-Fantastic-MiniMaxH3-PromptBuilder
Added: + - [ComfyUI-MiniMaxH3Mod
Added: + ](https://github.com/Luisacaotica/ComfyUI-MiniMaxH3Mod)
 
Removed: - ](https://github.com/Adudeguyman/ComfyUI-Fantastic-MiniMaxH3-PromptBuilder)
 
  ---

Before

- [ComfyUI-PlagueKind-Nodes](https://github.com/PlagueKind/Comfyui-PlagueKind-Nodes)

- [ComfyUI-KJNodes](https://github.com/kijai/ComfyUI-KJNodes)

- [Rgthree-comfy](https://github.com/rgthree/rgthree-comfy)

- [Nvidia_RTX_Nodes_ComfyUI](https://github.com/Comfy-Org/Nvidia_RTX_Nodes_ComfyUI)

- [ComfyUI-VideoHelperSuite](https://github.com/Kosinkadink/ComfyUI-VideoHelperSuite)

- [ComfyUI-H3-Motion-Context-MultiRef

](https://github.com/seitanism/ComfyUI-H3-Motion-Context-MultiRef)

- [ComfyUI-Fantastic-MiniMaxH3-PromptBuilder

](https://github.com/Adudeguyman/ComfyUI-Fantastic-MiniMaxH3-PromptBuilder)

---

After

- [ComfyUI-PlagueKind-Nodes](https://github.com/PlagueKind/Comfyui-PlagueKind-Nodes)
- [ComfyUI-KJNodes](https://github.com/kijai/ComfyUI-KJNodes)
- [Rgthree-comfy](https://github.com/rgthree/rgthree-comfy)
- [Nvidia_RTX_Nodes_ComfyUI](https://github.com/Comfy-Org/Nvidia_RTX_Nodes_ComfyUI)
- [ComfyUI-VideoHelperSuite](https://github.com/Kosinkadink/ComfyUI-VideoHelperSuite)
- [ComfyUI-H3-Motion-Context-MultiRef
](https://github.com/seitanism/ComfyUI-H3-Motion-Context-MultiRef)
- [ComfyUI-MiniMaxH3Mod
](https://github.com/Luisacaotica/ComfyUI-MiniMaxH3Mod)


---
Latent Upscaler Changed +1 −1

Diff

Removed: - [minimax_h3_latent_upscaler_3d_bf16.safetensors](https://huggingface.co/LBH-123-AI/Minimax_h3_latent_Upscaler/resolve/main/minimax_h3_latent_upscaler_3d_bf16.safetensors)
Added: + [minimax_h3_latent_upscaler_3d_bf16.safetensors](https://huggingface.co/LBH-123-AI/Minimax_h3_latent_Upscaler/resolve/main/minimax_h3_latent_upscaler_3d_conv_v1/minimax_h3_latent_upscaler_3d_conv_v1_bf16.safetensors)

Before

[minimax_h3_latent_upscaler_3d_bf16.safetensors](https://huggingface.co/LBH-123-AI/Minimax_h3_latent_Upscaler/resolve/main/minimax_h3_latent_upscaler_3d_bf16.safetensors)

After

[minimax_h3_latent_upscaler_3d_bf16.safetensors](https://huggingface.co/LBH-123-AI/Minimax_h3_latent_Upscaler/resolve/main/minimax_h3_latent_upscaler_3d_conv_v1/minimax_h3_latent_upscaler_3d_conv_v1_bf16.safetensors)
Optional Turbo Loras Changed +1 −0

Diff

⋯ 3 unchanged lines
 
  supporting file required. make the folder models/h3_adaln
 
Added: +
  [h3_silu_temb_grid.safetensors](https://huggingface.co/deAPI-ai/minimax-h3-33b-int8/resolve/main/loras/h3_silu_temb_grid.safetensors)
 
  ---

Before

[H3-PK-Parasyte-Turbo.safetensors](https://huggingface.co/Plaguekind/H3-Lora/resolve/main/H3-PK-Parasyte-Turbo.safetensors)

[fasth3_6step.safetensors](https://huggingface.co/Plaguekind/H3-Lora/resolve/main/otherturbo/fasth3_6step.safetensors)

supporting file required. make the folder models/h3_adaln

[h3_silu_temb_grid.safetensors](https://huggingface.co/deAPI-ai/minimax-h3-33b-int8/resolve/main/loras/h3_silu_temb_grid.safetensors)

---

After

[H3-PK-Parasyte-Turbo.safetensors](https://huggingface.co/Plaguekind/H3-Lora/resolve/main/H3-PK-Parasyte-Turbo.safetensors)

[fasth3_6step.safetensors](https://huggingface.co/Plaguekind/H3-Lora/resolve/main/otherturbo/fasth3_6step.safetensors)

supporting file required. make the folder models/h3_adaln


[h3_silu_temb_grid.safetensors](https://huggingface.co/deAPI-ai/minimax-h3-33b-int8/resolve/main/loras/h3_silu_temb_grid.safetensors)

---
V10 Notes Removed +0 −13

Diff

Removed: - - UPDATE NODEPACK TO V1.5.5 AND COMFYUI TO NEWEST VERSION, and requirements.
Removed: -
Removed: - - Added Refmod Support, create in subgraph.
Removed: -
Removed: - - Dropped turbo, it was ruining video and audio. default WF settings will get you 4 skips for "9" steps.
Removed: -
Removed: - - Added my H3 Combo node that does all modes.
Removed: -
Removed: - - Overall should be better quality, faster, and less overhead with reduced peak vram.
Removed: -
Removed: - - Use reference map to know what the LM is calling everything.
Removed: -
Removed: - - Shift node in Sampler subgraph for people who want to keep using turbo for some reason.

Before

- UPDATE NODEPACK TO V1.5.5 AND COMFYUI TO NEWEST VERSION, and requirements.

- Added Refmod Support, create in subgraph.

- Dropped turbo, it was ruining video and audio. default WF settings will get you 4 skips for "9" steps.

- Added my H3 Combo node that does all modes.

- Overall should be better quality, faster, and less overhead with reduced peak vram.

- Use reference map to know what the LM is calling everything.

- Shift node in Sampler subgraph for people who want to keep using turbo for some reason.

After

Not present in the later snapshot.

21 other sections unchanged.

Evidence · Observed by PinDiff

Earlier
Snapshot #498 · revision 4b5b724 · Recorded at 21 Sep 2026, 06:48 UTC
Later
Snapshot #534 · revision f1e1089 · Recorded at 22 Sep 2026, 06:39 UTC
Source
README.md at each revision, stored by PinDiff. Sections are compared individually.
Stored file SHA256
09391daa7ea8d7d1a963f9b6a3701ccbdeabc94391e2384d28b851ba4a08b59b → 5d667b4d5624179afef5b57ceefe61733053bdf5792655918fa1b37d328b59a2

Repository status

Changed

Repository revision moved

Earlier
Revision 4b5b724 (opens in a new tab) · 17 files · 1.29 MB
Later
Revision f1e1089 (opens in a new tab) · 18 files · 1.44 MB

A new revision does not by itself mean files changed; the categories above show what differs.

Evidence · Observed by PinDiff

Earlier
Snapshot #498 · revision 4b5b724 · Recorded at 21 Sep 2026, 06:48 UTC
Later
Snapshot #534 · revision f1e1089 · Recorded at 22 Sep 2026, 06:39 UTC
Source
Default-branch head commit observed via the Hugging Face API.

Files

Changed

3 other files changed

2 added 1 removed

Other files
FileStatusBeforeAfter
PlagueKind-MinimaxH3-V11.json ADDED — Git blob ID 509b5f53b3572c4e1b2cdd014e1b1d8fa7e4467f 146.0 KB
old/PlagueKind-MinimaxH3-V10.json ADDED — Git blob ID 8845f22dc6c78c9c0bf15e9d54b1b800ce2551ee 132.7 KB
PlagueKind-MinimaxH3-V10.json REMOVED Git blob ID 8845f22dc6c78c9c0bf15e9d54b1b800ce2551ee 132.7 KB —

Evidence · Observed by PinDiff

Earlier
Snapshot #498 · revision 4b5b724 · Recorded at 21 Sep 2026, 06:48 UTC
Later
Snapshot #534 · revision f1e1089 · Recorded at 22 Sep 2026, 06:39 UTC
Source
File hashes and sizes from the repository file manifest. Weights, licence, configuration, Python and model card files are covered above.

Evidence observed by PinDiff from the Hugging Face API and stored when recorded. PinDiff compares stored metadata, hashes and text files; it does not download weight files. Change events recorded between these snapshots