Providers

Orithos supports differential testing across LLM providers to compare how different models handle the same security probes.

Differential testing

Run the same probe set against multiple providers simultaneously to compare security postures. Differential results highlight which models are more susceptible to specific attack classes.

curl -X POST https://api.orithos.dev/v1/scans/differential   -H "Authorization: Bearer {api_key}"   -H "Content-Type: application/json"   -d '{
  "probe_set": "prompt-injection",
  "agent_ids": ["agent_openai", "agent_anthropic", "agent_google"]
}'

Parallel execution

Differential scans execute probes against all selected agents in parallel. Each agent receives the same probe payload simultaneously. Results are aggregated into a comparison view showing which agents passed, failed, or errored on each probe.

Supported providers