Embed Evidoc's Knowledge-Graph reasoning engine into your product, agent, or pipeline. Every answer comes with citations your users can verify.
No SDK. No auth dance. Just one HTTPS call, and you're integrated.
curl -X POST "https://graphrag-apim-wg3temevssbja.azure-api.net/graphrag/hybrid/query" \
-H "Content-Type: application/json" \
-H "Ocp-Apim-Subscription-Key: <YOUR_KEY>" \
-H "X-API-Version: v2" \
-d '{"query": "What is the contract termination notice period?"}'Returns a synthesized answer, citations to specific document passages, and the reasoning path used.
Single-doc Q&A, multi-hop reasoning across files, or wide-coverage comparison — pick the strategy per query, or let the server choose.
Every answer includes the exact passages it came from. Your users — or your compliance team — can verify the source in one click.
All traffic flows through Azure API Management. You hold a subscription key. We handle managed-identity tokens, throttling, and backend rotation.
Each response includes exact token usage and the model that synthesized the answer — attribute cost per query, per customer, per feature.
Dedicated subscription keys, isolated throughput, and Virtual Network integration available as add-ons for production deployments.
Use a fast/cheap model for high-volume queries, a frontier model for the ones that matter — overrideable per request.
Add contract Q&A to existing review workflows. Citations let lawyers verify every claim before signing off.
Answer tier-1 questions from a curated knowledge base. Citations let support agents audit responses before they reach the customer.
Surface policy citations alongside AI-generated risk summaries. Auditors get click-to-verify trails into the underlying policies.
Give your LLM agent a "search company documents" tool that returns cited facts instead of plausible-sounding inventions.
Buy prepaid credits and spend them as you go. No subscription, no monthly commitment, and credits never expire. Every API call draws down your balance based on real usage — a simple lookup costs a few credits, a complex multi-hop query costs more.
Credits are spent as you go — a typical query costs about 38 credits (billed by real usage, so simple queries cost less) and one page of document indexing costs 60 credits. Query figures are typical estimates; page figures are exact.
Start for $1Per-tenant subscription keys with isolated throughput, committed volume discounts, SLAs, and Virtual Network integration are available for production deployments. Let's understand your use case and quote what fits.
Endpoints, request/response schemas, error codes, Python and cURL examples — the same document we hand to integration customers.
Download PDF