hunyuan/text-to-video
Turn text prompts into high quality videos with Tencent Hunyuan Video.
This page covers the open-weights text-to-video path in the MiniMax H3 family. You describe visuals, camera moves, and sound in one prompt; the model returns an MP4 with matching stereo audio at 480p or 768p.
Use it when you want a lower-cost open-weights H3 route for prompt trials, storyboards, and short social drafts. Start at 5 seconds to lock composition and audio, then lengthen once the beat works. For the official MiniMax H3 text-to-video endpoint with different resolution tiers, see the related MiniMax H3 models on RunComfy.
Turn text prompts into high quality videos with Tencent Hunyuan Video.
Generates up to 4-minute songs with vocals from style tags and lyrics
Seedance 2.5 Reference 480p: Multi-reference draft video at lower cost
Generates natural speech and audio from text, reference audio, or an image
Create lifelike video motion fast with Seedance Pro for design pros
Transform static visuals into cinematic motion with Kling O1's precise scene control and lifelike generation.
MiniMax H3 Open is an open-weights text-to-video model that turns a written prompt into a short clip with picture and native stereo audio together. It fits prompt trials, storyboards, and social drafts when you want 480p or 768p output without a separate sound-design pass.
MiniMax H3 Open is the open-weights H3 text-to-video path with 480p and 768p output and per-second billing at those tiers. The official MiniMax H3 text-to-video model on RunComfy uses different resolution options (such as 768p and 2K). Pick MiniMax H3 Open for the open-weights route; use official MiniMax H3 when you need that product's resolution ladder.
MiniMax H3 Open supports duration from 5 to 15 seconds (default 5), resolution 480p or 768p, and aspect ratios 21:9, 16:9, 4:3, 1:1, 3:4, 9:16, and 9:21. Check the RunComfy parameter panel for the live controls on this page.
Yes. MiniMax H3 Open renders native stereo audio in the same pass as the picture—dialogue cues, effects, and ambience can be steered with an Audio: line in the prompt. If you omit audio direction, the model still supplies a soundtrack.
For MiniMax H3 Open, structure the prompt as style, a short timestamped timeline, explicit camera moves, and an Audio: line with entrance cues. Iterate at 5 seconds to lock composition and sound, then raise duration once the beat works. A short negative list (no text, no cuts) helps reduce drift.
Yes. Prototype in the RunComfy model UI, then call the same MiniMax H3 Open template through the RunComfy API with matching parameters. Generations consume usd / credits on both paths; scale volume once the prompt looks right.
MiniMax H3 Open is billed per second of generated video: $0.05 per second at 480p and $0.09 per second at 768p. A 5-second 480p clip costs about $0.25; longer durations scale linearly with the length you select.
Use 480p on MiniMax H3 Open for faster, cheaper drafts and batch variants. Switch to 768p when you need sharper finals for reviews, ads, or platforms that favor higher resolution.
RunComfy is the premier ComfyUI platform, offering ComfyUI online environment and services, along with ComfyUI workflows featuring stunning visuals. RunComfy also provides AI Models, enabling artists to harness the latest AI tools to create incredible art.
