Add today workflow imports

2026-08-13 18:07:56 +00:00
parent 3fffaa0ca0
commit 6657fd99fc
4 changed files with 4034 additions and 0 deletions
@@ -0,0 +1,56 @@
+# Dual Engine MiniMax-H3 + LTX-2.5
+
+## Summary
+
+One ComfyUI graph with a switch between a continuous `MiniMax-H3` chain and an `LTX-2.5` scene pass. It is a useful compare-and-contrast workflow for deciding whether a shot wants one long take or an edited cut structure.
+
+## Model Family
+
+- `minimax`
+- `ltx`
+
+## Status
+
+- `imported-only`
+
+## Source
+
+- Workflow repo: <https://huggingface.co/joeygambino/Dual-Engine-H3-LTX25>
+- Workflow file: <https://huggingface.co/joeygambino/Dual-Engine-H3-LTX25/resolve/main/workflows/DUAL_H3_LTX25.json?download=true>
+- Original publisher: `joeygambino`
+- Date imported: `2026-08-13`
+
+## Developer notes
+
+- The source README describes the graph as "one script, one switch".
+- The `MiniMax-H3` lane is for a single continuous take.
+- The `LTX-2.5` lane is for a scene that wants internal cuts.
+- The source README points `LTX-2.5` users to `joeygambino/LTX-2.5-Quantized` for the smaller builds.
+
+## Our notes
+
+- This is the most interesting same-day hybrid workflow hit because it is not just another model dump.
+- It is a better comparison tool than a first local run on an `RTX 3070 / 8 GB`; the full graph is more of a reference pack than a light test.
+- If you want to judge H3 against LTX on grammar and continuity, this is the cleanest daily find.
+
+## Required custom nodes
+
+- `ComfyUI-H3-Multishot`
+- `ComfyUI-Easy-Use`
+- `ComfyUI-kjnodes`
+- `ComfyUI-Impact-Pack`
+- `ComfyUI-Ollama`
+- `ComfyUI-GGUF` if you use the quantized LTX lane
+- `RES4LYF`
+- `WhatDreamsCost-ComfyUI`
+- `comfyui-videohelpersuite`
+
+## Required model files
+
+- `Comfy-Org/MiniMax-H3`
+- `joeygambino/LTX-2.5-Quantized` or an equivalent quantized `LTX-2.5` build for the second lane
+
+## Notes
+
+- Best treated as a benchmark pack for deciding which video grammar is worth pushing locally.
+- The H3 lane is the more direct fit if you care about continuous takes.
@@ -0,0 +1,50 @@
+# MiniMax H3 INT8 Reference-to-Video
+
+## Summary
+
+`MiniMax-H3` reference-to-video workflow with `INT8`, Motion Context, a dedicated reference LoRA path, and latent upscaling. This is the most practical same-day MiniMax consistency-oriented update.
+
+## Model Family
+
+- `minimax`
+
+## Status
+
+- `imported-only`
+
+## Source
+
+- Workflow repo: <https://huggingface.co/javawock7618/comfy-MiniMax-H3-workflows>
+- Workflow file: <https://huggingface.co/javawock7618/comfy-MiniMax-H3-workflows/resolve/main/MiniMax_int8_R2V-javano2608.14.json?download=true>
+- Original publisher: `javawock7618`
+- Date imported: `2026-08-13`
+
+## Developer notes
+
+- This is the `R2V` branch from the source pack.
+- The source README describes the pack as using `Sage-Attention`, `Spectrum`, `Lightx2v LoRA`, experimental `Turbo-LoRA`, `Motion Context`, and latent upscaling.
+- The `R2V` branch expects `minimax_h3_ref_lora_rank_256_bf16.safetensors`.
+
+## Our notes
+
+- This is the strongest today-only `MiniMax-H3` workflow for actual consistency and reference control.
+- It is still heavy for an `RTX 3070 / 8 GB`, but it is much more interesting than a novelty-only turbo drop.
+- If you want one MiniMax workflow to test against `LTX 2.3`, this is the one.
+
+## Required custom nodes
+
+- `ComfyUI-Spectrum-MiniMax-H3`
+- `ComfyUI-H3-Motion-Context`
+- `ComfyUI-MiniMaxH3_LatentUpscaler`
+
+## Required model files
+
+- `Comfy-Org/MiniMax-H3`
+- `minimax_h3_ref_lora_rank_256_bf16.safetensors` from `Kijai/MiniMax-H3-experimental`
+- optional `Kijai/MiniMax-H3_comfy`
+- optional `drbaph/MiniMax-H3-Turbo-Lora-ComfyUI`
+
+## Notes
+
+- This is the better `MiniMax-H3` pick when you care about reference fidelity, not just speed.
+- The pack also ships FLF and TTS workflows, but this import is the dedicated R2V graph.