Decision Workspace
jailguard vs llm-guard-ml vs vallum
Side-by-side comparison of Rust crates
47
jailguard
experimentalv0.1.2
Pure-Rust prompt-injection detector with 1.5MB embedded MLP classifier. 98.40% accuracy, p50 14ms CPU inference, 8-class attack taxonomy. Apache-2.0/MIT alternative to Rebuff and Lakera Guard.
40
llm-guard-ml
experimentalv0.1.0
ONNX-runtime-backed scanners for llm-guard. Catches paraphrased / novel prompt-injection attacks the rules tier can't. CPU by default; CUDA / CoreML / DirectML opt-in.
49
vallum
experimentalv0.8.16
Security boundary between AI coding agents and your shell — redacts secrets, neutralizes prompt injection, sanitizes untrusted terminal output, audits every command.
Core Metrics
| jailguard | llm-guard-ml | vallum | |
|---|---|---|---|
| Health Score | 47 | 40 | 49 |
| Total Downloads | 101 | 23 | 477 |
| 30d Downloads | 0 | 0 | 0 |
| Dependents | 0 | 0 | 0 |
| Releases | 3 | 1 | 25 |
| Last Updated | 148d ago | 145d ago | 68d ago |
| Age | 4m | 4m | 3m |
Health Breakdown
jailguard
Maintenance
12
Quality
12
Community
6
Popularity
2
Documentation
15
llm-guard-ml
Maintenance
6
Quality
15
Community
5
Popularity
2
Documentation
12
vallum
Maintenance
16
Quality
12
Community
6
Popularity
3
Documentation
12
Technical Details
| jailguard | llm-guard-ml | vallum | |
|---|---|---|---|
| Version | 0.1.2 | 0.1.0 | 0.8.16 |
| Stable (≥1.0) | ✗ No | ✗ No | ✗ No |
| License | MIT OR Apache-2.0 | MIT OR Apache-2.0 | MIT OR Apache-2.0 |
| Dependencies | 29 | 5 | 15 |
| Crate Size | 892KB | 23KB | 480KB |
| Features | 10 | 4 | 1 |
| Yanked % | 0.0% | 0.0% | 0.0% |
| Edition | 2021 | 2024 | 2021 |
| MSRV | 1.85 | 1.85 | 1.85 |
| Owners | 1 | 1 | 1 |
Links
Quick Verdict
- •vallum leads with a health score of 49/100, but none of the options score above 80.