Deploy with Helm
The chart deploys the guardrail service and, optionally, its Redis and Presidio dependencies.
Install
helm install pseudonymizer \
./deploy/helm/palena-litellm-pseudonymizer \
--namespace guardrails --create-namespaceBy default this brings up the guardrail plus a single-instance Redis and both Presidio services — a self-contained stack for a proof of concept. Helm prints the in-cluster URL and a ready-to-paste LiteLLM guardrail block on install.
Production: bring your own dependencies
Disable the bundled Redis and Presidio and point at managed instances:
redis:
enabled: false
url: redis://my-managed-redis:6379/0
presidio:
analyzer:
enabled: false
url: http://my-presidio-analyzer:3000
imageRedactor:
enabled: false
url: http://my-image-redactor:3000NER service
With config.detectionEngine: ner (the default) the chart also deploys the NER service, ghcr.io/palenaai/palena-ner:<appVersion>: GLiNER2-PII on CPU with the model baked into the image, so it needs no network access at runtime. It takes about 2 Gi of memory (limit 3 Gi) and 10–30 s to load; /health answers 503 until then, and the guardrail's /readyz waits for it. Presidio then only runs its pattern recognizers, in config.language, so the analyzer stays on the stock English image whatever config.languages says.
ner:
replicas: 2
torchThreads: 2 # match resources.limits.cpu
resources:
requests: { cpu: "1", memory: 2Gi }
limits: { cpu: "2", memory: 3Gi }To use your own NER service, set ner.enabled: false and ner.url. config.detectionEngine: presidio keeps the previous behaviour (spaCy per language, no NER service).
Organization names
Add your known company names for high-precision masking (see Organization detection):
presidio:
analyzer:
organizations:
- Novartis
- RocheCommon values
| Key | Default | Purpose |
|---|---|---|
replicaCount | 2 | Guardrail replicas. |
image.tag | chart appVersion | Service image tag. |
config.detectionEngine | ner | ner (NER service for names) or presidio (spaCy per language). |
config.entities | PERSON | Detected entity types. |
config.languages | [] (= config.language, en) | presidio engine only: languages to analyze in; several switch to the multilingual analyzer image. |
config.entityScoreThreshold | PHONE_NUMBER:0.4 | Per-type minimum score. |
config.allowList | "" | Terms never pseudonymized. |
config.allowListBuiltin | true | Add the built-in product/tool name list. |
ner.threshold | 0.9 | Minimum span confidence. |
ner.replicas / ner.parallel | 1 / 0 (= replicas) | NER pods and batches in flight per request. |
ner.cache.entries / ner.cache.ttl | 500000 / 1h | Cache of lines already analyzed, per pod (~100 MiB full). |
ner.chunkChars / ner.chunkOverlap | 1000 / 150 | Window size for long lines. |
apiKey.enabled | false | Require an x-api-key header. |
redis.enabled | true | Bundle Redis. |
presidio.analyzer.nerOrganization | true | spaCy org detection. |
presidio.analyzer.organizations | [] | Exact-match org deny-list. |
presidio.analyzer.denyLists | [] | Exact-match deny-lists for any entity type. |
presidio.analyzer.multilingualImage | ghcr.io/palenaai/palena-presidio-analyzer:<appVersion> | Analyzer used for anything but English alone. |
autoscaling.enabled | false | HPA on CPU. |
serviceMonitor.enabled | false | Prometheus Operator scrape. |
The full list lives in the chart's values.yaml.
Security
Set an API key before exposing the service beyond the proxy:
apiKey:
enabled: true
value: "a-long-random-secret" # or reference an existing Secret
# existingSecret: my-secret
# existingSecretKey: api-keyUninstall
helm uninstall pseudonymizer --namespace guardrails