Benchmarking SDXL ConvRot INT8: HSWQ Quantization Balances Quality and VRAM
Zestyclose_Bake3680 · reddit · 2026-08-02
A developer shared benchmark results for their independently developed HSWQ (Hybrid-Sensitivity-Weighted Quantization) technology applied to SDXL models using ConvRot INT8.
The tests compared FP16 references against quantized outputs:
- Overall Advantage: SDXL is inherently well-suited for ConvRot INT8. By adding FP16 protection to key layers, HSWQ generally achieves higher SSIM scores and lower MSE than native quantization.
- Bias Correction: Enabling bias correction tends to improve scores, although the effectiveness varies across different models.
- Exceptions: In a few rare cases (e.g., perfectionAsianILXL), native ConvRot INT8 slightly outperforms the HSWQ approach.
More from Infra
- Baidu open-sources FlashMask: up to 1.55x faster than FlashAttention-4 for long-context training — BaiduResearch · 2026-10-16
- GYRO quant of Qwen3.8-Flash-Next tested: near Q3 quality at much smaller size — PlasticRevenue4601 · 2026-10-16
- RL train-sample systems are frontier AI's hardest infra problem — and a trade secret — brianryhuang · 2026-10-16
- Local tuning of GLM-5.3-Flash on 2×GB10 lifts decode from 49 to 58 tok/s — EAccelerate_42 · 2026-10-16
- 3x DGX Spark cluster runs GLM-5.3-Flash locally behind a proxy, 1.75M tokens/day at 119W — Teknium · 2026-10-16
- Optuna creator Akiba: when 2 hyperparameters cost $20M, what's left for BO/HPO — Stefania_druga · 2026-10-16