⚡ Request for SVDQuant Checkpoints Support & Development of FLUX fn4 with Nunchaku Technology

**Description**:
This issue proposes adding support for **SVDQuant checkpoints** and developing a custom **FLUX fn4 checkpoint** using Nunchaku’s SVDQuant, a 4-bit quantization technique. SVDQuant reduces memory and increases efficiency by absorbing outliers in model weights via low-rank components, providing a robust solution for memory-constrained environments without degrading performance.

**Justification**:
SVDQuant uniquely addresses the challenge of memory and latency in large models:
- **Efficient Memory Management**: By absorbing weight outliers, SVDQuant reduces memory usage by up to 3.5× in models like FLUX.1, achieving 3× speedup over weight-only quantized models.
- **Maintained Quality**: Visual fidelity is preserved, and SVDQuant matches 16-bit model quality, even with 4-bit quantization, making it ideal for high-performance applications.

**Implementation Steps**:
1. **Load SVDQuant Checkpoints**: Integrate support for the loading of SVDQuant-formatted checkpoints.
2. **Develop FLUX fn4 Checkpoint**: Train and validate FLUX fn4 using SVDQuant’s quantization, with benchmarks against non-quantized models to ensure quality retention.
3. **Optimize Performance**: Use Nunchaku’s kernel fusion to minimize data movement and reduce latency by combining low-rank and low-bit processing.

**References**:
- [Project](https://github.com/mit-han-lab/nunchaku) | [Paper](https://arxiv.org/abs/2409.04429) | [Blog](https://hanlab.mit.edu/projects/svdquant) | [Demo](https://svdquant.mit.edu/)

No repository files, tests, or entry points are identified in the issue. Start by locating the checkpoint-loading path and the existing FLUX integration, then define how SVDQuant checkpoints and the proposed FLUX fn4 checkpoint would be validated against non-quantized models and measured for performance.

Written by the indexing model from the issue text.