In this guide
Short answer: Generate video and audio from a prompt using the Pika adapter on MiniMax H3. The listed files total about 42.04 GB on disk. The repositories list MiniMax H3 Community License Agreement.
Output example
This workflow was run on a PC with an RTX 3090. We downloaded the saved output and inspected sampled video frames. It is one prompt and seed, not a general quality benchmark.
Audio stream checks confirm technical presence only. We have not verified sound quality, speech accuracy or lip sync by listening.
Prompt used for this example
integrated_multimodal_description: [Shot 1] Cinematic macro photography of a small cobalt-blue ceramic cup on a dark walnut table beside a rain-streaked window. A thin curl of steam rises from the hot tea, folds into a soft spiral, then disperses. Rain droplets slide slowly down the glass behind it, catching warm amber reflections from a distant streetlamp. The camera makes a very slow small push toward the cup, keeping the rim sharp and the rainy street beautifully defocused. The cup stays planted, the tea remains inside it, and the warm side lighting stays consistent. One continuous five-second shot, no cuts, no people, no lettering.
overall_soundscape: Soft rain pattering against glass, a quiet indoor room tone, and a faint distant passing car.
non_diegetic_music: N/AObserved limits: One fixture does not exercise optional inputs or establish general reliability. Audio stream presence does not establish speech accuracy, sound quality or lip sync.
The featured image documents the workflow graph. The output example above comes from a separate completed run, with its recorded prompt and limitations.
Download and load the workflow
Drag the downloaded JSON file onto the ComfyUI canvas to load it.
Install the model files and any custom nodes listed below. If the workflow uses input images, video or audio, select your own files in the input nodes before running it.
Which files go where
Place each file under your ComfyUI directory in the folder shown. A text encoder turns the prompt into model conditioning. A VAE decodes the sampled representation into image or audio data. A LoRA is an adapter loaded alongside the base model.
| Part | File | Size | Folder |
|---|---|---|---|
| vae | minimax_h3_video_vae_int8_convrot.safetensors | 2.81 GB | models/vae |
| vae | minimax_h3_audio_vae_fp32.safetensors | 0.61 GB | models/vae |
| diffusion_models | minimax_h3_fl2va_pruned_int8_convrot.safetensors | 20.97 GB | models/diffusion_models |
| text_encoders | qwen3vl_32b_minimax_h3_nvfp4_awq.safetensors | 15.69 GB | models/text_encoders |
| loras | pika_h3_flex_fl2va_v1_bf16.safetensors | 1.96 GB | models/loras |
The listed files total about 42.04 GB on disk. Disk size is not a measured VRAM requirement. Reuse matching files already installed rather than downloading a second copy.
Settings in this workflow
| Control | Saved value |
|---|---|
| Sampler | lcm |
| LoRA strength | 1.0 |
| Sigma intervals | 6 |
| Duration / frame rate | 5 seconds / 24 fps |
| Saved image-to-video node size | 1344 × 768 |
| Resolution selector | 16:9, 0.98 megapixels, 16-pixel alignment |
Saved sigma schedule: 1.0, 0.972973, 0.923077, 0.8, 0.571429, 0.333333, 0.0.
How to use it
- Open the text-to-video template and replace the example prompt with your scene.
- Describe the visible action and sound. The example separates scene description, overall_soundscape and non_diegetic_music.
- Keep the fl2va checkpoint and Pika fl2va adapter paired, with lcm and the saved sigma schedule.
- Choose duration and size, then generate when ready and inspect the saved video and audio.
What to check before changing the graph
The outer graph has no image input. Its subgraph is named Image to Video (Pika H3 Flex), but the saved wiring uses it for text-to-video. Do not add a reference image merely to match that label. The seven sigma values define six sampling intervals. Duration is converted to the model’s frame grid, so the saved inner frame widget is not the final duration control. Use the separate reference-to-video workflow when you need picture inputs.
Licence
The repositories list MiniMax H3 Community License Agreement. This is a community licence with conditions. Read the current terms before commercial use; a public download does not mean unrestricted use. See our model licence tracker for the wider collection.
Troubleshooting
| Problem | Likely cause | Fix |
|---|---|---|
| Missing nodes | Your installation differs from the captured environment | Update ComfyUI and reload the template before reconnecting nodes. |
| Model missing from a dropdown | File is in the wrong folder or the list is stale | Check the exact filename and folder above, then refresh the model list. |
| Unexpected output size or length | A connected input overrides an internal widget | Trace the size and duration connections from the outer graph. |
Where to go next
FAQ
Are these output examples?
The featured image shows the workflow graph. The Output example section shows the actual saved generation from a PC with an RTX 3090.
Can I change the saved settings?
Yes. Keep the documented configuration as a reference and change one control at a time. Samplers, adapters and noise schedules work together, so treat a replacement recipe as a separate configuration.
Sources
- Comfy-Org/MiniMax-H3: model files and repository licence metadata.
Keep building
Get the cheat sheets and planners or browse all ComfyUI workflows.
Need help building a production workflow? Talk to Automate Digital.