LLM Guard
A scanner suite for input and output. The fastest thing to put in front of an app.
93
sovereigntyLLM Guard from Protect AI is a collection of composable scanners covering the practical threat surface: prompt injection, jailbreak attempts, personal data, toxicity, secrets in prompts, code detection, relevance and refusal detection on output. You choose which scanners to run and in what order, and it sits as a layer in front of and behind the model. MIT licensed, self-hosted, and the quickest of these to add to something already running.
OPEN SOURCEMITSELF-HOSTLOCAL-FIRST
LicenseMIT
PricingFree, MIT licensed.
Open sourceYes
Self-hostableYes
Local-first dataYes
What it does well
- +Broad scanner set covering both input and output threats
- +Compose only the checks you need — each is independent
- +Straightforward to insert into an existing application
- +MIT, fully self-hosted
Where it falls short
- −Model-based scanners need their own compute
- −Every added scanner adds latency
- −Thresholds require real tuning to avoid false positives
LLM Guard as an alternative to
Where LLM Guard shows up in our comparisons, and how it ranked.
LLM Guard head-to-head
Straight comparisons against the tools people weigh it against.