Since the movie is an anthology of different stories animated by different studios, I let this LoRa be influenced by a combination of all the syles to hopefully create it's own somewhat unique style while still giving a general nod and homage to this great movie.
VERSIONS:
H3 Minimax
Trigger: HvyMtl Style
Strength 1
Example prompts:
1) Desert Hoverbike Pilot
integrated_multimodal_description:
8-second video in HvyMtl Style. An adult red-haired pilot sits astride a battered hoverbike at a remote desert fueling station, wearing a torn-open flight jacket over a tiny armored halter, very low-slung utility pants, thigh straps, heavy boots, and a weapons belt packed with ammunition and blades. At 0.00 seconds, she sits sideways on the bike with one boot planted in the sand, jacket hanging open as hot wind whips her hair across her face. From 0.60 to 2.60 seconds, she flicks open a metal lighter, lights a cigarette, and slowly exhales while watching the empty horizon. From 2.60 to 4.80 seconds, distant shapes appear through the blowing dust and several strange travelers at the station turn nervously toward them. From 4.80 to 6.60 seconds, rusted alien machinery behind her shudders awake and the hoverbike engine begins vibrating beneath her. From 6.60 to 8.00 seconds, she drops the cigarette into the sand, swings fully onto the bike, arches forward over the controls, and launches away as dust and sparks explode behind her.
overall_soundscape:
Hot desert wind, sputtering fuel machinery, distant alien voices, lighter snap, cigarette inhale, hoverbike engine growl, sand blasting across metal.
non_diegetic_music:
Driving heavy-metal track with distorted guitar riffs, pounding drums, dirty bass, and a rising guitar lead during the launch.
2) Barbarian Queen Above the Lava Canyon
integrated_multimodal_description:
8-second dark fantasy video in HvyMtl Style. An adult blonde barbarian queen stands on a black stone balcony above a vast lava canyon, wearing a jeweled metal battle harness, ornate bracers, a tiny low-slung ceremonial loincloth, thigh chains, and heavy leather boots. A massive sword rests beside her while hooded worshippers gather far below between towering demonic pillars. At 0.00 seconds, she stands in profile near the balcony edge, one hand resting on the sword pommel while molten light moves across her body. From 0.60 to 2.80 seconds, the camera pushes closer as hot wind lifts her hair and the hanging chains at her hips. From 2.80 to 4.80 seconds, the worshippers below raise their arms while the canyon begins shaking and streams of lava surge brighter. From 4.80 to 6.60 seconds, she slowly draws the enormous sword free and turns toward the camera, blade glowing orange from the reflected heat. From 6.60 to 8.00 seconds, lava erupts violently behind her as she plants the sword beside her and holds a commanding full-body pose against the blast.
overall_soundscape:
Roaring lava, deep canyon wind, distant chanting, stone rumbling, metal jewelry shifting, heavy sword scraping free.
non_diegetic_music:
Slow crushing heavy metal with thunderous toms, distorted bass, sustained guitar notes, and a huge riff when the lava erupts.
KREA2
Tigger: HvyMtl Style.
Strength 1 should work but if you get a more modern output, increase to 1.1 or 1.2
Z Image Turbo -
Strength of .7-.9 seem to work best. "HeavyMetalZit style of" <your prompt> will get the style working for you.
Hunyuan Video
This was created using an existing database of just 512x512 images I had on hand using Diffusion Pipe.
example prompt: HeavyMetalstyle animation of a green alien with a long face, skinny features, orange vest, tan shirt, dark green pants is sitting against the control panel of his spaceship. Lots of switches and light panels and a monitor in the background. The alien is looking at the viewer and has his left hand raised.
SDXL
Based on the popular animation styles used in Heavy Metal. Might need to increase weight to 1.2 if using a realistic model. Also helps to use tags like flat art, or 2D flat art. My first attempt at this so be nice.
Flux
Not perfect by any means but it's a good starting point. Hopefully we'll be able to train loras locally with 12gb sooner than later and I can spend more time on it. I tagged the training data in natural language but it seems to respond well to both types of prompts. Normal guidance seems to work fine. The model can get a little horny sometimes with the boobies but so did the movie so it's fitting.



















