Felldude/CLIP-L & CLIP-G Full FP32 (Zer0Int & Simulacrum)

This is a full restoration of Zero0Int and AbstractPila work on the CLIP-L model. In 1000's of test images the FP32 model corrected bad anatomy in cases where the FP16 model failed (Seed to Seed) In most cases the images have little difference but when differences exist the FP32 model was reliably more accurate (As it should be mathematically) Note the vision models use the VIT-L vision blocks Zer0Int has a new finetune of the VIT/Vision model here MIT License Copyright (c) 2021 OpenAI Permission is hereby granted, free of charge, to any person obtaining a copy of this software and associated

<iframe src="https://silytics.si/embed/item/06f4aa1c-5d59-47e2-8c82-63127281d84e" width="420" height="190" style="border:0" loading="lazy" title="Silytics entry"></iframe>

Silytics shows public metadata from Civitai and links to the original. Information may be incomplete or out of date. Last refreshed 17 October 2026. Licenses and terms of the original apply to the work itself. Not financial, legal or professional advice.