<h3 id="!-if-you-like-this-model-please-remember-to-give-me-a-like!">如果你喜欢本模型, 请记得给我点赞哦! If you like this model, please remember to give me a like!</h3><h1 id="or-update-notes:">模型更新内容 | Update notes:</h1><h2 id="26-10-16">26-10-16</h2><ul><li>新版本Fn-Moment Anima-29B-V5，基于 <a target="_blank" rel="ugc" href="https://civarchive.com/models/2855007/anima-29b">Anima-29B </a>模型，继承V4.1模型风格，获得更多角色提示词支持。</li><li>由于我认为自己仍然没有完全掌握2.9B的使用方式，因此新版本更接近测试版。</li><li>本次例图依旧是添加 turbo lora 后生成，不添加 turbo lora 情况下推荐使用 dpmpp_2m_sde_gpu+beta1_1, step: 25, cfg: 4</li><li>！如使用 Anima 非 2.9B 的 turbo lora，需要配合 <a target="_blank" rel="ugc" href="https://github.com/sparklingcoffee777/ComfyUI-Anima-2.9B-loraPatch">Anima-2.9B-loraPatch</a> 插件使用；或者直接使用 2.9B 的 turbo lora <a target="_blank" rel="ugc" href="https://civarchive.com/models/2857451/anima-29b-turbo-lora-remapped?modelVersionId=3227488">Anima-29b-turbo-lora-remapped </a>！</li><li>The new version, Fn-Moment Anima-29B-V5, is based on the Anima-29B model and inherits the style of V4.1, with support for more character prompts.</li><li>Since I feel I still haven’t fully mastered how to use 2.9B, this new version is closer to a beta release.</li><li>The example images this time were still generated with the turbo lora added. If you are not using the turbo lora, it is recommended to use dpmpp_2m_sde_gpu + beta1_1, steps: 25, cfg: 4.</li><li>！If you are using an Anima turbo LoRA that is not the 2.9B version, it must be used together with the <a target="_blank" rel="ugc" href="https://github.com/sparklingcoffee777/ComfyUI-Anima-2.9B-loraPatch">Anima-2.9B-loraPatch</a> plugin; Or just use the 2.9B turbo LoRA, <a target="_blank" rel="ugc" href="https://civarchive.com/models/2857451/anima-29b-turbo-lora-remapped?modelVersionId=3227488">Anima-29b-turbo-lora-remapped</a> ！</li></ul><h2 id="26-08-28">26-08-28</h2><ul><li>新版本 Fn‑Moment Anima‑V4.1，进一步强化画面电影质感，提升背景画面稳定性。</li><li>优化丰满体型人物的呈现效果，提升皮肤质感，完善人物肢体表现。</li><li>画面洁净度进一步提升，可根据使用需求加载 detailer lora 以强化细节效果。</li><li>The new version Fn‑Moment Anima‑V4.1 further enhances the cinematic texture of visuals and improves the stability of background images.</li><li>Optimizes the rendering effect of characters with fuller body shapes, elevates skin texture, and refines character limb performance.</li><li>Visual cleanliness is further improved. The detailer lora can be loaded according to usage requirements to strengthen detail effects.</li></ul><h2 id="26-08-23">26-08-23</h2><ul><li>新增模型 Fn‑Moment Anima‑V3.0‑Fix，该模型针对 Fn‑Moment Anima‑V3.0 在未使用 turbo Lora 情况下的生成效果进行优化调整，进一步提升画面输出稳定性。</li><li>Added the new model Fn‑Moment Anima‑V3.0‑Fix. This model optimizes and adjusts the generation performance of Fn‑Moment Anima‑V3.0 when the turbo Lora is not used, further improving the stability of image output.</li></ul><h2 id="26-08-21">26-08-21</h2><ul><li>新版本模型 Fn-Moment Anima-V3.0，针对大量用户想要非 turbo 版本的需求开发。样图虽用了 turbo LoRA 生成，但实测该模型脱离 turbo LoRA 依旧可以产出高质量图像.</li><li>新版本模型 Fn-Moment Anima-V3.0，侧重电影感风格, 角色更加丰满, 画风更有张力.</li><li>推荐使用turbo LoRA获得与例图相同的效果.</li><li>如果你喜欢我的模型，不妨点个赞支持一下；如果愿意分享你创作的作品，或是提出对模型的建议，那就再好不过了. 感谢大家！</li><li>The new‑version model Fn‑Moment Anima‑V3.0 is developed in response to numerous users' demand for a non‑turbo version. Although the sample images were generated with the turbo LoRA, practical tests show that this model can still produce high‑quality images without the turbo LoRA.• The new‑version model Fn‑Moment Anima‑V3.0 focuses on a cinematic style, delivering more well‑rounded characters and greater stylistic intensity.• It is recommended to use the turbo LoRA to achieve the same effects as the sample images.• If you like my model, please give it a like to show your support. It would be wonderful if you could share your created works or offer suggestions about the model. Thank you all!</li></ul><h1 id="or-introduction:">模型介绍 | Introduction:</h1><ul><li>本模型基于 Anima BASE1.0 模型, 融合多个个人训练LoRa以及Tubro LoRa调整生成.</li><li>本模型强化了光影表现, 角色相比 Fn-Mix Anima-Turbo 更加丰满, 画风更有张力.</li><li>本模型不需要负面提示词.</li><li>当前所有测试皆基于Comfyui, 未在Webui中测试.</li><li>This model is based on the Anima BASE1.0 model, generated by integrating multiple personally trained LoRAs for adjustment.</li><li>This model enhances light and shadow performance; the characters are fuller compared to Fn-Mix Anima-Turbo, and the art style has more tension.</li><li>This model does not require negative prompts.</li><li>All current tests are conducted based on Comfyui and have not been tested in Webui.</li></ul><h1 id="or-usage-tips:">使用技巧 | Usage tips:</h1><ul><li>可挑选以下tag添加以强化风格 | You may select the following tags to enhance the style: chromatic aberration, depth of field, photo background, blurry background, blurry foreground, lens flare, bokeh</li><li>可挑选以下tag添加以生成边框 | You may select the following tags to generate borders: polaroid photo, white border, black border</li><li>注意其中部分tag相互冲突, 请不要一次性添加所有tag! | Note that some of these tags conflict with each other, so please do not add all tags at once!</li></ul><h1 id="or-recommend">参数推荐 | Recommend：</h1><ul><li>Resolution: 832*1216/1024*1536</li><li>Steps: 12/30(V3.0 no turbo)</li><li>CFG: 1/4-5(V3.0 no turbo)</li><li>Sampler: euler_ancestral/euler(V3.0 no turbo)</li><li>Scheduler: simple/nomal(V3.0 no turbo)</li><li>Hires Fix Step: 5</li><li>Hires Fix CFG: 1/5(V3.0 no turbo)</li><li>Hires Fix Sample: euler</li><li>Hires Fix Scheduler: nomal</li><li>Hires Fix Denoise: 0.3</li></ul>

Fn-Moment Anima-Turbo

Fn-Moment-Anima-29B-V5.safetensors

Files with SHA256: 46a3d554be35f3a91b4cad452ec41a23c04868ba1fe623af9845e03965ec886f