This release adds 3 notable features for engineering teams evaluating rollout.
✓ No known CVEs patched in this version
Topics
+14 more
Summary
AI summaryReplaces PyTorch + sentence-transformers with ONNX Runtime inference, reducing install size from ~2.5GB to ~200MB while preserving embeddings and behavior.
Full changelog
What's Changed
Replaces PyTorch + sentence-transformers with direct ONNX Runtime inference. Install drops from ~2.5GB to ~200MB. Same model, same 768-dimension embeddings, same ChromaDB collections — zero user-facing change.
Highlights
- Zero new dependencies —
onnxruntime,tokenizers,huggingface-hub,numpyare all already installed viachromadb - Identical embeddings — cosine similarity vs PyTorch: 1.0000000, multilingual verified across 8 languages
- Optimized model — uses
model_O4.onnx(529MB, half the unoptimized size), identical latency roampal doctor— now checks new deps, hints if torch is installed but no longer needed
Install / Upgrade
pip install --upgrade roampal
Weekly OSS security release digest.
The CVE patches and breaking changes that affected production tools this week. One email, every Sunday.
No spam, unsubscribe anytime.
Share this release
About roampal-ai/roampal-core
Outcome-based persistent memory for AI coding tools. Memories that help get promoted, memories that mislead get demoted. Works with Claude Code and OpenCode via hooks + MCP.
Related context
Beta — feedback welcome: [email protected]