"The AWP series 3D large-scale model is now available. This model's character is derived from the "Duiyou" open-source Avatar under Alibaba's umbrella. Characters generated possess distinct recognizability. This model can be aptly used for crafting 3D icons and 3D scenes. It is highly recommended for projects relating to 3D character and IP training within LoRA.Instructions:<ol><li>For direct outputs in t2i, selectable portrait types include: Close-up, Upper body, and Full body. The ADetailer plugin is required to fine-tune facial features with a weight range of 0.25-0.4.</li><li>Expressions like "smile," "happy," "angry," and "stick out tongue" are available. Additional expressions can also be introduced via the ADetailer plugin.</li><li>For a broader range of Prompts, kindly refer to the provided sample output images.</li><li>It's recommended to employ DPM++ 2M Karras for optimal sampling.</li></ol>Disclaimer: You are solely responsible for maintaining the content generated based on this model service and using it following your independent judgment. Ensure compliance with all relevant laws, regulations, and the terms of this service. Any intellectual property issues arising from the generated content based on this service must be addressed by you. I bear no responsibility for any consequential damages or losses. Without prior written consent, this model service cannot be integrated into any third-party software or service. Furthermore, users should not disclose the connection between the generated content and this service to any third party, whether directly or indirectly, expressly, or implicitly. Violations of this agreement may result in claims for damages on the part of the author.

AWPoly3D

awpoly3d_v10.safetensors

v1.0

AWpoly3d

The large 3D model of the **AWPoly3D 1.0** AWP series has finally arrived. The characters in this model were trained based on Alibaba's “Duiyou” open source Avatar as a data set, and the generated characters are very recognizable. You can also use it to create 3D icons and 3D scenes. It is highly recommended to use this model to train Lora related to 3D characters and IPs.

**Instructions for use:**
When the T2i is straight out, portrait scenes are optional, such as: Close-up, Upper Body, Full Body. Straight out requires the ADetailer plug-in to control the face (weight 0.25-0.4). You can choose emoticons such as: smile, happy, angry, stick out tongue, etc. (can also be added to ADetailer). More Prompts references, see the example directly. Sampling recommendation DPM++.

**2M Karras Disclaimer:** Text generated by you based on this model service. The content is maintained by you and used at your own discretion. You must use it under the conditions of compliance with applicable laws and regulations and these terms of service. Any intellectual property issues arising from content generated from this service are handled by you. You are responsible for the generated content, and I am not responsible for any losses caused by this. Except as otherwise provided by laws and regulations, this model service may not be integrated into any third-party software or service, and the relationship between generated content and this service may not be disclosed to any other third party by direct, indirect, explicit, or implied means; non-individual users (including but not limited to enterprises and for-profit organizations) may not use the model for commercial purposes (including but not limited to deepfakes, memes, face-swapping, etc.) without my written authorization and consent, and you have the right to recover damages caused to the author himself by violating this agreement.

Files with SHA256: 3cde97066290f3e461a2be226eec2ee2f39b07a19a3701b8e17ead58ca63f923