UPSTREAM PR #1169: feat: prioritize gguf and safetensors formats for embeddings and LoRAs

[loci-dev]

Mirrored from leejet/stable-diffusion.cpp#1169

Models available as multiple file types were likely converted from one another, so we should prioritize native, safer and more efficient formats.

[loci-review]

Explore the complete analysis inside the Version Insights

I've generated a summary report for your project. Here are the key findings:

Performance Summary Report

Project: stable-diffusion.cpp (auroralabs-loci)
Pull Request: #8

Key Results:

• ✅ No significant performance regressions detected
• ✅ No significant performance improvements detected
• All function-level performance metrics remain within ±2% variance threshold

Conclusion:

Pull Request #8 is performance-neutral, meaning the changes maintain stable performance characteristics without introducing any measurable performance degradation or improvement. This is a positive outcome, particularly if the PR includes new features, bug fixes, or code improvements.

The comparison between the base version (aaa4da61-e62e-11f0-81f2-dbb430499cb5) and target version (54538151-e8ec-11f0-81f2-dbb430499cb5) shows that both response time and throughput metrics for all modified functions remain stable.