Thelocallab/Pika-H3-Flex-ComfyUI-Workflow

Pika H3 Flex - ComfyUI reference workflow (tested settings)

A ComfyUI workflow for Pika H3 Flex by Pika Labs, on MiniMax H3: one to three pictures and a shot-by-shot prompt in, a short scene with camera moves, dialogue, sound and music out, in six sampling steps. It runs locally and is not the pika.art app. This repo holds only the workflow, no model weights.

🎬 Video tutorial: https://youtu.be/_CUlPWpRa8o

File

• Pika-H3-Flex-References-SemanticBridge.json: drag it into ComfyUI. Notes inside list every model link and the settings.

Tested settings (already set)

Sampler lcm, Pika's six manual sigmas (1.0, 0.972973, 0.923077, 0.8, 0.571429, 0.333333, 0.0), CFG 1, adapter strength 1, VSA gate sparsity 0.75, Semantic Bridge strength 0.15. Keep the memory cleanup nodes on.

Measured (RTX A4500, 20 GB, one reference, gate on)

• 5 s at 0.6 MP (1056x608): 1:43 sampling, 2:11 with the gate off
• 15 s at 0.6 MP: about 7:00 sampling, about 11:15 for the whole run with the 2x upscale
• 16 and 12 GB, simulated by capping VRAM on the 20 GB card, one reference: a 5 s clip at 0.6 MP stayed inside both caps at the same speed. A 15 s clip finished but peaked at about 17 GB, past both caps, so start with short clips on those cards.
• 0.8 MP at 15 s ran out of memory on 20 GB. 1344x768 at 15 s with three references ran out on 20 GB and finished on a 24 GB card.

Models

• minimax_h3_ref2va_pruned_int8_convrot.safetensors (about 21 GB), diffusion_models: https://huggingface.co/Comfy-Org/MiniMax-H3/tree/main/diffusion_models
• qwen3vl_32b_minimax_h3_nvfp4_awq.safetensors (about 16 GB), text_encoders: https://huggingface.co/Comfy-Org/MiniMax-H3/tree/main/text_encoders
• minimax_h3_video_vae_fp16.safetensors (about 5.2 GB) + minimax_h3_audio_vae_fp32.safetensors (about 600 MB), vae: https://huggingface.co/Comfy-Org/MiniMax-H3/tree/main/vae
• pika_h3_flex_ref2va_v1_bf16.safetensors (about 2 GB), loras: https://huggingface.co/Comfy-Org/MiniMax-H3/tree/main/loras
• fasth3_vsa_gate.safetensors (about 3.9 GB), loras: https://huggingface.co/barelymining/ComfyUI-MiniMax-H3-FastVideo/tree/main
• MiniMaxH3_SemanticBridge_v1.safetensors (about 11 MB), create models/semantic_bridge: https://huggingface.co/speach1sdef178/MiniMax-H3-Semantic-Bridge/tree/main
• About 48 GB in total. Official model page: https://huggingface.co/Pika-Labs/Pika-H3-Flex

Custom nodes

• Ref2VA VSA by Kablex: https://github.com/Kablex/ComfyUI-Ref2VA-VSA
• MiniMax H3 Semantic Bridge by speach1sdef178: https://github.com/Speach1sdef178/MiniMax-H3-Semantic-Bridge
• Memory Cleanup, Nvidia RTX Nodes, KJNodes, rgthree, VideoHelperSuite: ComfyUI Manager, Install Missing Custom Nodes

Use PyTorch built for CUDA 13 (cu130) with an NVIDIA driver 580 or newer.

Prompt tips

• Make each clip one complete little story in one location, then cut between clips.
• Write positions the way they appear on screen ("the left of the frame"), and say where the camera is in every shot.
• Write out why things happen. "The gap is too narrow for the drone's rotor arms" gave a crash you can actually follow.
• Use things the model already knows. For something invented, give it a reference picture.
• With several references, give each picture its own definition, and use only the appearance of a character sheet, without the panels.
• Test your prompt on a short clip first.

Credits

Pika Labs (Pika H3 Flex), MiniMax (H3), the FastVideo team and Kablex (VSA gate), speach1sdef178 (Semantic Bridge).

Free guide: https://www.patreon.com/TheLocalLab/posts/174460405

Also on Civitai: https://civitai.com/models/3038436

Licence

The Pika H3 Flex adapters and MiniMax H3 are under the MiniMax H3 Community License Agreement. Read it before commercial use.