Add MiniMax H3 Upscaler with model links and video comparison

Removed upstream alternatives for upscalers and LoRAs.
This commit is contained in:
SOLRICKS
2026-11-05 18:13:44 +03:00
committed by GitHub
parent 1a3c1da13c
commit 3d1afebb72
-2
View File
@@ -53,8 +53,6 @@ Download the models and example workflows from [SOLRICKS/MiniMax-H3-Upscaler](ht
| Experimental INT8 refiner LoRA | [h3_upscale_refiner_int8.safetensors](https://huggingface.co/SOLRICKS/MiniMax-H3-Upscaler/resolve/main/h3_upscale_refiner_int8.safetensors?download=true) | `ComfyUI/models/loras/` |
| FP16 latent upscaler | [minimax_h3_latent_upscaler_3d_conv_v1_fp16.safetensors](https://huggingface.co/SOLRICKS/MiniMax-H3-Upscaler/resolve/main/minimax_h3_latent_upscaler_3d_conv_v1_fp16.safetensors?download=true) | `ComfyUI/models/latent_upscale_models/` |
Upstream alternatives: [FP16 latent upscaler](https://huggingface.co/LBH-123-AI/Minimax_h3_latent_Upscaler/tree/main/minimax_h3_latent_upscaler_3d_conv_v1) and [BF16 refiner LoRA](https://huggingface.co/sandpies/ComfyUI-H3-Video-Upsampler/blob/main/lora/open_h3_regenerate_2k_r256pp_bf16.safetensors). The node also accepts the original BF16 refiner filename.
You also need a compatible **MiniMax H3 FL2VA diffusion model** and **H3 video VAE**, such as the files in [Kijai/MiniMax-H3-experimental](https://huggingface.co/Kijai/MiniMax-H3-experimental). Put these in `models/diffusion_models/` and `models/vae/` respectively. Connect a model without generation/Turbo/style LoRAs: this node applies the refiner LoRA itself.
### Workflow downloads