A lora to celebrate the science fiction movies and shows of the 1950s. It's fairly narrowly trained, doing well with space suits, bulky robots and moon-like landscapes.All images/videos in showcase contain ComfyUI workflows.Hunyuan 1.5:First try. All examples done without lightx, 20 steps.Wan 2.2:The Low noise lora has absolutely most effect for style. High noise still affects the outcome, but to a much less degree.For the Wan showcase, I use 25 steps: 15/10. 0.8 strength, and dpm++_sde scheduler in ComfyUI. I haven't had a lot of time for testing, but it might be that it loses strength as you go up in frames.I use Q4.gguf for High noise, and Q6 for Low noise, for no particular reason. No block swapping.Wan lora trained with diffusion-pipe.

1950s Scifi Movies Lora

I've reused a lot of prompts from the Hunyuan version for you to compare.I'm a bit unsure about this version. Does it need more training? Sometimes it looks great. Sometimes the style is right, but the quality is too good for being 50s. Like Wan wants to "fix" it. I'm experimenting a bit by removing the negative prompt.

50s_scifi_wan_low_30.safetensors

Wan 2.2 T2V 14B Low

Flux.2 Klein 9B Base v1.0

v1.0 - Hunyuan Video 1.5

v1.0 - ZImage(De)Turbo

Wan 2.2 T2V 14B High

v1.0 - Comfy compatible

v1.0 - Diffusers Lora

A lora to celebrate the science fiction movies and shows of the 1950s. It's fairly narrowly trained, doing well with space suits, bulky robots and moon-like landscapes.I'm still learning about training, and one thing I learned is that since I've trained at a width of max 544p, the further you deviate from this, the less the lora will affect the output.I should vary and increase the size more for the next version.Hunyuan loras trained using <a target="_blank" rel="ugc" href="https://github.com/a-r-r-o-w/finetrainers">https://github.com/a-r-r-o-w/finetrainers</a><a target="_blank" rel="ugc" href="https://github.com/comfyanonymous/ComfyUI/pull/6174￼But"> </a>All images/videos in showcase contain ComfyUI workflows.Wan 2.2:The Low noise lora has absolutely most effect for style. High noise still affects the outcome, but to a much less degree.For the Wan showcase, I use 25 steps: 15/10. 0.8 strength, and dpm++_sde scheduler in ComfyUI. I haven't had a lot of time for testing, but it might be that it loses strength as you go up in frames.I use Q4.gguf for High noise, and Q6 for Low noise, for no particular reason. No block swapping.Wan lora trained with diffusion-pipe.

&lt;p&gt;I&#39;ve reused a lot of prompts from the Hunyuan version for you to compare.&lt;/p&gt;&lt;p&gt;I&#39;m a bit unsure about this version. Does it need more training? Sometimes it looks great. Sometimes the style is right, but the quality is too good for being 50s. Like Wan wants to &#34;fix&#34; it. I&#39;m experimenting a bit by removing the negative prompt.&lt;/p&gt;

Files with SHA256: 9503d48d385e9fc576055be993d394e7e1af18a288649025fbf7fc283b0a50c8