MiniMax H3 gets a one-click ComfyUI RunPod template, with real speed benchmarks
TheLocalLab · reddit · 2026-08-18
A community one-click ComfyUI RunPod template for MiniMax H3 auto-downloads all models and supports first/last frame, image-to-video, and text-to-video; a 4090 or 5090 is recommended. Rough timings on a 4070 Ti Super (20 steps, SageAttention, first/last frame): 5s @ 0.5MP ≈ 7 s/it, 10s 9:16 ≈ 17 s/it, 15s ≈ 31 s/it, 10s @ 1MP ≈ 52 s/it, and 15s @ 0.8MP reaches 72–118 s/it — cost scales sharply with resolution and duration, so start small.
More from Infra
- Implementing Hybrid AI on Apple with Dynamic Profiles — rudrank · 2026-08-18
- Native vLLM on Windows 11 for AMD RX 6000: One-Click Installer — Dizzy_Counter2481 · 2026-08-18
- Developers discuss Gemini API billing structure and reconciliation — Neat-Ad-4224 · 2026-08-18
- AWS Bedrock Adds Cross-Region Inference for OpenAI Models — DigitalColmer · 2026-08-18
- Neoclouds Rise as AI Demands Synchronized Compute, Not Just Chips — DavidLinthicum · 2026-08-18
- Hobbyist trains a diffusion model on a 264KB-RAM microcontroller, ~70s per image — PandaBean18 · 2026-08-18