Artifact Risk Predictor (Arctenox's Essentials)

ArtifactRiskPredictor

🔍 Artifact Risk Predictor - Analyze Quality Before Decoding Analyzes latent space for signs of common generation artifacts: • Overbaked faces (too many steps/high CFG) • Melted hands (high-frequency noise patterns) • Flat contrast (collapsed dynamic range) • Noisy/grainy results (sampling artifacts) • Oversaturated regions (CFG overshoot) ⚡ What This Node Does: Provides INFORMATIONAL risk assessment - does not stop or control workflow. Use the risk_score output with switches, routers, or conditional nodes to build self-correcting workflows that can retry or adjust parameters. 🎯 Risk Score Output: • 0.0-0.2: ✅ Low risk - probably good • 0.2-0.4: ⚠️ Mild risk - watch for issues • 0.4-0.6: 🟡 Moderate risk - likely problems • 0.6-0.8: 🟠 High risk - expect artifacts • 0.8-1.0: 🔴 Critical risk - probably unusable 💡 Example Use Cases: • Pass risk_score to a switch node to conditionally decode • Log warnings for quality monitoring in batch generation • Use with retry nodes to regenerate high-risk latents • Build adaptive workflows that adjust CFG based on risk 🔧 Detection Methods: • Latent statistics (mean, std, range) • Frequency analysis (detect noise patterns) • Dynamic range checks (contrast collapse) • Parameter correlation (overbaking detection) • Channel balance analysis ⚙️ Risk Thresholds: • Conservative: Flag more potential issues (fewer false negatives) • Balanced: Standard detection (good default) • Permissive: Only flag obvious problems (fewer false positives) ℹ️ Note: This node analyzes and reports - it does not prevent decoding. Combine with other nodes to create conditional/adaptive workflows.

Pack: Arctenoxs-Essentials_ComfyUI

custom_nodes.Arctenoxs-Essentials_ComfyUI