High-fidelity Hybrid ConvRot NVFP4 quantization for Z-Image Turbo diffusion models. Built from a complete native ConvRot INT8 UNet via the reverse method (converting lowest-impact layers to NVFP4 in ascending order of trajectory impact). This is highly useful for users who need to strictly manage their VRAM resources (~53–58% savings) while maintaining high image fidelity and completely eliminating trajectory bifurcations.

Technical details: https://github.com/ussoewwin/Hybrid-Sensitivity-Weighted-Quantization

How to quantize (Z-Image Hybrid NVFP4): md/How to quantize Z Image - Hybrid NVFP4.md

ComfyUI Loader for ConvRot NVFP4: To use these models in ComfyUI, please use this custom node: ComfyUI-HSWQ-Loader-and-Tools

Z Image ConvRot NVFP4 Benchmark Test Results (published tables): benchmark result/benchmark_zi_nvfp4.md

Benchmark (Deterministic Latent Trajectory Divergence)

Multi-seed trajectory evaluation (20 seeds per model, 200 total evaluations across 10 Z-Image models):

Metric	HSWQ ConvRot Hybrid NVFP4	Native NVFP4 (Full Model)	Advantage
Final Cosine (Avg)	0.960 – 0.974	0.900 – 0.947	+0.020 to +0.060 higher fidelity
Bifurcation Rate	0/200 (0.0%)	3/200 (1.5%)	Zero bifurcations across all seeds
Latent MSE Reduction	0.74 – 1.25	1.45 – 2.54	36% – 61% lower quantization error
VRAM Footprint	~53% – 58% Savings	~50% Savings	Significantly reduced VRAM usage

📦 Available Models

Filename	Base Model	Version	License
`moodyProMix_zitV13_hswq_hybrid_nv80_convrot_nvfp4.safetensors`	Moody Pro Mix	zit v1.3 (nv80)	CreativeML Open RAIL++-M
`moodyProMix_collectorsEdition_hswq_hybrid_nv90_convrot_nvfp4.safetensors`	Moody Pro Mix	Collector's Edition (nv90/nv100)	CreativeML Open RAIL++-M
`moodyRealMix_zitV7_hswq_hybrid_nv100_convrot_nvfp4.safetensors`	Moody Real Mix	zit v7.0 (nv100)	CreativeML Open RAIL++-M
`moodyRealMix_xhsEdition_hswq_hybrid_nv110_convrot_nvfp4.safetensors`	Moody Real Mix	XHS Edition (nv100/nv110)	CreativeML Open RAIL++-M
`darkBeast30BF16INT8_dbzit9DIMRclaw_hswq_hybrid_nv100_convrot_nvfp4.safetensors`	Dark Beast	dbzit9 DIMRclaw (nv100)	CreativeML Open RAIL++-M
`unstableRevolution_V3Fp16_hswq_hybrid_nv90_convrot_nvfp4.safetensors`	Unstable Revolution	v3.0 FP16 (nv90/nv100)	CreativeML Open RAIL++-M
`gonzalomoZpop_insta2_hswq_hybrid_nv80_convrot_nvfp4.safetensors`	gonzalomo Z-Pop	insta2 (nv80/nv100)	CreativeML Open RAIL++-M
`divingZImageTurbo_v70Fp16_hswq_hybrid_nv90_convrot_nvfp4.safetensors`	Diving Z-Image Turbo	v7.0 FP16 (nv90/nv99)	CreativeML Open RAIL++-M
`beyondREALITY_V30_hswq_hybrid_nv100_convrot_nvfp4.safetensors`	Beyond REALITY	v3.0 (nv100)	CreativeML Open RAIL++-M
`2127ZImageAsianUtopian_v40Turbo_hswq_hybrid_nv100_convrot_nvfp4.safetensors`	2127 Z-Image Asian Utopian	v4.0 Turbo (nv100)	CreativeML Open RAIL++-M
`copaxTimeless_xplusZ13_hswq_hybrid_nv100_convrot_nvfp4.safetensors`	Copax Timeless	xplus Z13 (nv100)	CreativeML Open RAIL++-M
`unstablebastard_v14_hswq_hybrid_nv100_convrot_nvfp4.safetensors`	Unstable Bastard	v1.4 (nv100)	CreativeML Open RAIL++-M
`zimageTurboByStable_2602BF16_hswq_hybrid_nv100_convrot_nvfp4.safetensors`	Z-Image Turbo	2602 BF16 (nv99/nv100)	CreativeML Open RAIL++-M

📜 Credits & License

🏆 Special Acknowledgement

We extend our deepest respect and gratitude to the Nunchaku Team for their groundbreaking work on SVDQ quantization and for sharing their models with the community. This collection relies heavily on their research and original implementation.

Original Repository: nunchaku-tech/nunchaku-sdxl

Base Models

These models are derivatives of their respective creators. All credit for aesthetic tuning and model training belongs to the original creators.

Moody Pro Mix / Moody Real Mix / Dark Beast: Created by catlover1937.
Unstable Revolution / Unstable Bastard: Created by Yamer.
gonzalomo Z-Pop: Created by gonzalomo.
Diving Z-Image Turbo: Created by Diving.
Beyond REALITY: Created by Lykon.
2127 Z-Image Asian Utopian: Created by 2127.
Copax Timeless: Created by Copax.
Z-Image Turbo: Created by Tongyi-MAAS / Civitai Community.

Disclaimer: These models are provided for optimization and research purposes. Please adhere to the original licenses of the base models.

ussoewwin/Hybrid-Sensitivity-Weighted-Quantization-Z-Image-Hybrid-ConvRot-NVFP4

Hybrid-Sensitivity-Weighted-Quantization (HSWQ)

Benchmark (Deterministic Latent Trajectory Divergence)

📦 Available Models

📜 Credits & License

🏆 Special Acknowledgement

Base Models

ussoewwin/
Hybrid-Sensitivity-Weighted-Quantization-Z-Image-Hybrid-ConvRot-NVFP4