[MiniMax H3] NSFW I2VA / T2VA / FL2VA / R2VA Workflows 👾  Qwen3.5 Auto Prompt | ⚡ MiniMax-H3 Turbo LoRA | Native Audio | TensorRT Upscale | RIFE Interpolation

⚡ MiniMax-H3 Turbo LoRA (Optional — Faster &amp; Sharper)A distilled 4–8 step LoRA for MiniMax-H3 that replaces the default ~20-step sampling, with a dedicated ComfyUI node.<ul><li>Custom node: <a target="_blank" rel="ugc" href="https://github.com/Larryvrh/ComfyUI-MiniMax-H3-Turbo">Larryvrh/ComfyUI-MiniMax-H3-Turbo</a></li><li>Recommended checkpoint: <code>minimax_h3_turbo_v4_step600_ema.safetensors</code> (~744 MB)</li><li>Download: <a target="_blank" rel="ugc" href="https://huggingface.co/larryvrh/MiniMax-H3-Turbo-Lora/resolve/main/minimax_h3_turbo_v4_step600_ema.safetensors">larryvrh/MiniMax-H3-Turbo-Lora</a></li><li>Install: place the <code>.safetensors</code> in <code>ComfyUI/models/loras/</code></li><li>Usage: insert <code>MiniMax-H3 Turbo LoRA</code> between <code>Load Diffusion Model</code> and <code>SamplerCustomAdvanced</code>, and use <code>MiniMax-H3 Turbo Sampler</code> with scheduler <code>simple</code> at 6–8 steps.</li></ul>Works with all tasks: T2VA, I2VA, FL2VA and R2VA.

MinimaxH3NSFWI2VAT2VAFL2VAR2VAWorkflows_OneclickMMH3Turbo.zip

Files with SHA256: 0fe4bc05478fceab3ea1d8dbe074373a1706c41701548f6a8705e375c5fb22aa