IronLeo78 MTV (v0.0.2)
Minimax MTV: loader + MiniMax H3 reference-to-video + sampling + decode + Create Video + Save Video, in one node. Models: Turbo + custom LoRA; diffusion model, attention backend, Qwen3-VL text encoder and the 2 VAEs are hardcoded (v2.3: the optional 'models' input takes a Model Loader node that can replace them, and the step counts). Media: video (segment by segment) + up to three images + custom audio (whole song, or only its Vocals / Guitars / Drums / Bass: Audio Stem; the stems are separated once per song and saved to output/<Song>/stems/). Sampling is hardcoded (BasicGuider, res_multistep, simple, denoise 1.0); Turbo ON = 4 steps, Turbo OFF = 25 steps. The noise seed (default 1978) is set on the Model Loader node. Audio Latent: always Bypass Minimax (hardcoded): the audio you give (from the video or custom) is kept as is, the model never generates audio. Save Latent ON: the sampled latent (video + audio .latent files) and its conditioning (.conditioning) are saved next to the clip for a later refine / upscale in another workflow. OFF (default): only the clip is saved. Files are tagged <Song>_seg<segment>_A (Song = the Custom Audio File name, if one is selected) and logged in a manifest CSV. No outputs, no saved-video preview (the live latent preview stays): the clips are saved to the output folder and the run report is shown in the node. v2.5: segment panel (count / next / Reset), progress in the report, a quiet stop after the last segment, and a <clip>.json sidecar with the full prompt next to every clip. v2.6: jump to any segment (number field + Go / Latest in the panel), and the Guide can read the previous segment's last frame back from its clip. v2.8: Scene 2 / prompt_2 / Picture4-6 are gone (one clip per Run). by IronLeo78
- models
Inputs (16)
| Name | Type | Default | Description |
|---|---|---|---|
| Turbo | BOOLEAN | true | ON: Turbo LoRA applied, 4 steps. OFF: no Turbo LoRA, 25 steps (BasicGuider, no CFG, like the official r2v template). |
| Use Video | BOOLEAN | false | ON = v2v: the video is loaded and is the reference for megapixels/width/height. OFF = i2v: no frames are loaded, Picture1 is the reference. Exception: with Bypassed Audio Latent = Audio From Video, a selected Video File still supplies its soundtrack (audio only). Set Video File to None for silence. |
| Video File | COMBO | None | 1 options: None |
| Custom Audio File | COMBO | 1 options: None | |
| Audio Stem | COMBO | Song | Which part of the Custom Audio File is used. Song = the whole mix. Vocals / Guitars / Drums / Bass = only that stem (Guitars is Demucs' 'other' stem: guitars + keys + synths). The first time a stem is chosen for a NEW song, Hybrid Demucs separates all four (about 13 s for a 2 minute song) and saves them to output/<Song>/stems/<Song>_<Stem>.wav. After that every Run just reads the file. Needs a Custom Audio File. |
| megapixels | FLOAT | 0.20.1–1 | Target size. Use Video ON: the video sets the aspect ratio and width/height. Use Video OFF: Picture1 does. Pictures 2-3 always keep their own aspect ratio. |
| duration | FLOAT | 15.000–10000 | Duration in seconds. Also sets the video segment length when Use Video is ON. durationS = this value unchanged; durationF = Minimax frame count. |
| Next Segment On Run | BOOLEAN | false | ON: every Run moves on to the next segment (video frames, and/or the custom audio chunk when Bypassed Audio Latent = Custom Audio). OFF: always the first segment. |
| prompt | STRING | Prompt. Refer to references as <Picture i> / <Video 1> / <Audio j>. Numbers count only what is actually sent: an empty picture slot is skipped (Picture1 + Picture3 filled = <Picture 1> and <Picture 2>). Audio order: video soundtrack first (if any), then custom audio. The status box lists the exact mapping. | |
| Bypassed Audio Latent | BOOLEAN | false | Which audio is used (and stepped through). This audio is encoded into the audio latent (Bypass Minimax is always on). Audio From Video = the video soundtrack (silence if none). Custom Audio = the Custom Audio File (silence if none). It also decides which audio Next Segment On Run moves through: Custom Audio = the song is cut into chunks as long as the real video segment (frames / fps: duration 5 s = 124 frames = 5.17 s), one chunk per Run (works with Use Video OFF too); the segment counter at the top of the node decides which chunk is next (Jump). Audio From Video = the video soundtrack follows its frames, the custom audio is then used whole. |
| Save Latent | BOOLEAN | false | ON: besides the clip, the sampled latent (video + audio .latent files) and the conditioning (.conditioning: prompt + pictures, no guide) are saved in the song folder, next to the clip and sharing its name, so you can refine / upscale it later in another workflow. OFF (default): only the clip is saved. |
| Guide | BOOLEAN | true | Only works with Next Segment On Run = ON. ON: the last frame of the previous run is used as a guide for the first frame of this segment, so the clips continue each other. The first segment (or a run with no saved frame from the segment before) runs without a guide. OFF: no guide. |
| picture_1 | COMBO | None | 2 options: None, example.png |
| picture_2 | COMBO | None | 2 options: None, example.png |
| picture_3 | COMBO | None | 2 options: None, example.png |
| modelsopt | IRONLEO78_MTV_MODELS | Optional. Plug in the IronLeo78 MTV Model Loader to replace the hardcoded diffusion model, Turbo LoRA, text encoder, VAEs and step counts. Nothing connected = the hardcoded defaults. |
Outputs (0)
No outputs