NSFW Content filtering model offers vision based filtering solution to find explicit content. Its main use case is to check outputs of the generative models like Stable Diffusion. Model combines zero-shot capabilities of pretrained CLIP with a linear probing approach for fine tuning.
During fine-tuning, a light classification layer is trained on top of the frozen CLIP encoder, significantly improving the quality of detecting NSFW Content compared to just zero-shot classification.
Feature |
Training |
Inference |
|---|---|---|
| Data parallelism | Yes | N/A |
| Tensor parallelism | N/A | N/A |
| Pipeline parallelism | N/A | N/A |
| Sequence parallelism | N/A | N/A |
| Activation checkpointing | Yes (Uniform or Block) | No |
| FP32/TF32 | Yes | Yes (FP16 enabled by default) |
| AMP/FP16 | Yes | Yes |
| AMP/BF16 | Yes | Yes |
| BF16 O2 | Yes | No |
| TransformerEngine/FP8 | No | No |
| Multi-GPU | Yes | Yes |
| Multi-Node | Yes | Yes |
| Inference deployment | N/A | NVIDIA Triton supported |
| SW stack support | Slurm DeepOps/Base Command Manager/Base Command Platform | Slurm DeepOps/Base Command Manager/Base Command Platform |
| NVfuser | No | N/A |
| Distributed Optimizer | N/A | N/A |
| TorchInductor | No | N/A |
| Flash Attention | No | N/A |