<h1 id="anima-llm-layerwise-(fp8-fp32)">Anima LLM LayerWise (FP8-FP32)</h1>Use this in place of Qwen3_06B it is loaded as CLIP<h3 id="fp8-version-tends-to-smooth-detail-compared-to-fp32-or-bf16fp16">FP8 VERSION TENDS TO SMOOTH DETAIL COMPARED TO FP32 or BF16/FP16</h3><ul><li>Anima Artist and Character triggers fully working</li><li>Full FP32 Training - Adam not 8BIT</li><li>Uncensored LLM can be used for story or text generation</li><li>FP8 Model has best of both worlds with 50% weight reduction and FP32 blocks where it counts.</li></ul><hr />Comfy UI users --fp32-text-enc or --bf16-text-enc can be placed in your launch script to force BF16 or FP32 casting of the LLMNote: Top rated images where used for prompts as I do not know Anima well.

Qwen 06B Uncensored (Anima)

Files with SHA256: 018bfb16dbb2187dc6b5a0c05575558ab69a8a5d00e38391b8f95e98b0f83882