Help: Krea2 INT8 Quantization Workflow Yields Poor Results
derTommygun · reddit · 2026-07-06
A user is seeking help with the Krea2 Convrot INT8 quantization workflow. On an RTX 4070, they achieved satisfactory results using BF16/FP8 with lora (about one image per minute). However, while Convrot INT8 is faster, it shows a noticeable degradation in details, human anatomy, and overall quality.
They noted that others in the community have shared high-quality INT8 results, so they want to know what they are doing wrong and are asking for a working workflow or full settings (model, VAE, steps, sampler, resolution, etc.).
More from Infra
- Resin lets an agent build standalone LLM inference engines; vLLM and SGLang optimize DeepSeek kernels differently — AccBalanced · 2026-10-27
- Local benchmark: Flint matches Qwen3.6 quality at 50/52 while running 3.6x faster on M4 Pro — RichardSocher · 2026-10-27
- Anthropic S-1: CSP marketplace revenue jumped from 11% to 48% in three years — AccBalanced · 2026-10-27
- Alchemy adds effect/Workflow support across Durable Objects, Workflows and Lambda Durable Functions — samgoodwin89 · 2026-10-27
- Alchemy's Lambda Web Functions run plain Node web servers on Lambda, no OpenNext needed — samgoodwin89 · 2026-10-27
- Alchemy Cuts Cloudflare Bindings CPU Time by Up to 166× Via Profiling — samgoodwin89 · 2026-10-27