ComfyUI_MiniMax_H3_Extender

ComfyUI MiniMax H3 Extender

A ComfyUI custom node for MiniMax H3 designed to generate long, continuous video sequences from multiple clips while preserving motion, visual continuity, and audio continuity between generations.

The node combines Ref2VA conditioning, Motion Context, disk caching, multi-clip generation, image references, audio references, and final video/audio decoding into a much simpler workflow.

---

Project status

This is the final feature release of MiniMax H3 Extender.

Apart from possible bug fixes or compatibility corrections, I do not currently plan to continue active development of this project.

---

🚀 New: v3.0.5 — Latent Refine / Upscale

A new optional Latent Refine / Upscale mode is now available, inspired by the latent refine work introduced in PR #83.

When enabled, each clip is generated normally, then its latent is upscaled and passed through a second H3 refine pass before continuing through the Extender’s existing cache, Motion Context, preview and export pipeline.

There is no separate draft/refine workflow: the refined result becomes the clip’s final result directly.

Refine settings include:

• Scale Factor
• Refine Steps
• Refine Denoise
• automatic display of the resulting final video resolution

The same H3 model is reused for both passes, and the required latent upscaler model is downloaded automatically on first use.

---

🆕 v2.9.0 — Continue an Existing Video

MiniMax H3 Extender can now continue from an existing video.

• Connect a video to the new optional continue_existing_video input.
• The source becomes a locked Clip 0 and is automatically included at the beginning of the preview and final export.
• Clip 0 can be resized directly from its card using Original, Auto / MP, or Manual working resolution.
• The source video is normalized to the Extender working format without modifying the original file.
• The final frames of Clip 0 are used as the visual context for the first generated clip.
• Choose Ref lets you pick any frame from the source video and use it as a normal <Picture N> reference, useful for preserving character identity when the end of the source video does not show the subject clearly.
• Existing Motion Context behavior between generated clips remains unchanged.

This makes it possible to take an existing video and continue building it directly inside the Extender.

---

🆕 v2.8.0 — Optional individual Full Batch clip export

Full Batch can now optionally save each final clip as a separate video file using the new Save Individual Clips option in Final Decode / Preview.

The option is disabled by default, so the existing Full Batch workflow remains unchanged unless explicitly enabled.

When enabled, individual clips are exported only after all normal processing is complete. They use the same final video segments as the assembled output, including seam correction, per-clip color adjustments and FL2VA visible-frame trimming/handoffs.

Audio is captured during the same final PCM assembly pass used for the complete video, so no additional audio reconstruction pass is required.

No additional sampling or VideoVAE decode is introduced. Individual video streams are reused with stream-copy and saved in a dedicated folder beside the final assembled video.

Supported in:

• Ref2VA Motion ON
• Ref2VA Motion OFF
• FL2VA

---

🆕 v2.7.7 — Full Batch Autosave

• Optional automatic .ext project save after a successful Full Batch
• Supports Ref2VA Motion ON/OFF and FL2VA
• Preserves the actual generation seed and project settings
• Clip-by-Clip behavior remains unchanged

---

🆕 v2.7.5 — New Project button

Added a new New Project button to quickly start from a completely clean state.

It clears all current project data and cached references, removes prompts, resets the timeline to a single empty clip, and wipes previous computed/validated state — while keeping all global node se