| Name | Modified | Size | Downloads / Week |
|---|---|---|---|
| Parent folder | |||
| llmslim-0.3.1-py3-none-any.whl | 2026-08-13 | 64.5 kB | |
| llmslim-0.3.1-py3-none-any.whl.publish.attestation | 2026-08-13 | 9.7 kB | |
| llmslim-0.3.1.tar.gz | 2026-08-13 | 1.2 MB | |
| llmslim-0.3.1.tar.gz.publish.attestation | 2026-08-13 | 9.5 kB | |
| LLMSlim v0.3.1 source code.tar.gz | 2026-08-13 | 2.9 MB | |
| LLMSlim v0.3.1 source code.zip | 2026-08-13 | 3.0 MB | |
| README.md | 2026-08-13 | 836 Bytes | |
| Totals: 7 Items | 7.3 MB | 0 | |
Highlights
- Adds provenance-aware
ContextRolehandling so untrusted RAG, tool, and assistant content cannot gain protected priority solely from imperative wording. - Adds caller-supplied rewrite providers and validated rewrite/hybrid workflows.
- Improves token counting, inline-code handling, CJK behavior, and compression selection correctness.
- Makes the benchmark runner report actual pytest outcomes.
Verification
- 432 tests passed with 0 failures
- 92.57% branch coverage
- Ruff passed
- Benchmark suite passed on 43 bundled samples across 8 dataset files
Notes
v0.3.1 is a Python release. npm, Rust, ONNX, and WASM runtimes are not shipped. Rewrite and hybrid modes require a caller-supplied provider.
See the repository release notes and docs/release/v0.3.1-release-report.md for details.