The AI FinOps control plane

Govern AI spend in the request path. Budget gates, per-call cost metadata, and Security & Compliance — with 35+ providers behind a single API key.

CoreValue Dashboard
Requests+12%
12,481/ 1h
Cost−18%
$24.81today
Avg latency−8%
1.4sp50
ModelEndpointLatencyTokensCost
gpt-4o-mini/chat240ms2,104$0.003
claude-sonnet-4/chat1.2s1,840$0.028
gpt-4o/summarize3.4s3,218$0.048
gemini-2.0-flash/extract180ms612$0.000
claude-haiku-4/classify680ms984$0.002
1,000free requests/mo
35+providers
1-lineintegration
Per-callcost in every response
One line to integrate

Point your base URL at CoreValue. That's it.

Keep the OpenAI SDK you already use. Swap the baseURL and instantly get logging, routing, fallbacks, and per-call cost metadata across every provider — no SDK rewrite, no data pipeline to maintain.

  • Works with OpenAI, Anthropic, Gemini, Azure, Bedrock, Ollama & more
  • JS/TS, Python, cURL, and frameworks like LangChain & Vercel AI SDK
  • Automatic fallbacks & intelligent routing across providers
import OpenAI from "openai";

const client = new OpenAI({
  baseURL: "https://ai-gateway.corevalue.dev",
  apiKey: process.env.COVA_API_KEY,
});

const response = await client.chat.completions.create({
  model: "gpt-4o-mini", // claude-sonnet-4, gemini-2.5-flash, or 35+ providers
  messages: [{ role: "user", content: "Hello!" }],
});

Everything you need to ship LLM features

From the gateway that routes your requests to the FinOps controls that govern your spend — CoreValue covers the full lifecycle.

35+providers

AI Gateway

One API key unlocks 35+ providers with pass-through routing and automatic fallbacks. Native SDK support — no format translation required.

Every responsecost in headers

Per-call FinOps

Cova-Request-Cost and Cova-Credits-Remaining in every response header. The @cova/gateway SDK parses them into typed metadata.

Pre-executionspend control

Budget Gates

Block or downgrade requests before they reach the provider. Only an in-path proxy can govern spend before it happens.

Per call, per userchargeback-ready

Cost Tracking

Track cost by user, team, and feature. Chargeback-ready cost attribution with Cova-User-Id and Cova-Property headers.

Dual-limitersliding + token bucket

Rate Limiting

Sliding window + token bucket rate limiting with cost-based units. Per-key, per-org, per-segment controls.

PII scanningAES-GCM encryption

Security & Compliance

PII scanning & masking, AES-GCM encrypted cache, RBAC, and GDPR rights documented in our privacy policy.

Auto-retryordered fallback

Fallback Chains

Ordered fallback across providers and models. If your primary provider is down, CoreValue routes to the next automatically.

At-rest encryptiondual-key rotation

Encrypted Cache

Cache LLM responses with AES-GCM at-rest encryption and dual-key rotation. Your cached data is encrypted at rest, not stored in plaintext.

1-clickexecutive reports

ROI Scorecard

Generate ROI scorecards and executive reports that show the business impact of your AI spend.

Per-call FinOps

See spend per call — not per billing cycle

Every response includes Cova-Request-Cost and Cova-Credits-Remaining headers. The @cova/gateway SDK parses them into typed metadata. No other gateway exposes per-call FinOps metadata.

  • res.gateway.requestCost exact cost per request
  • res.gateway.creditsRemaining real-time credit balance
  • res.gateway.provider / res.gateway.model attribution metadata
  • res.gateway.gatewayMode passthrough or translated
fetchNative · @cova/gateway
per-call
import { fetchNative } from "@cova/gateway";
 
const { response, gateway } = await fetchNative(
  "https://ai-gateway.corevalue.dev/v1/chat/completions",
  {
    method: "POST",
    headers: {
      Authorization: `Bearer ${process.env.COVA_API_KEY}`,
      "Content-Type": "application/json",
    },
    body: JSON.stringify({
      model: "claude-sonnet-4",
      messages: [{ role: "user", content: "Hello" }],
    }),
    cova: { userId: "user-123" },
  }
);
 
console.log(gateway.requestCost);      // 0.0123
console.log(gateway.creditsRemaining); // 47.32
console.log(gateway.provider);         // "anthropic"
requestCost parsed·creditsRemaining: $47.32·provider: anthropic

Works with the tools you already use

35+ inference providers through a single OpenAI-compatible API. Plus the @cova/gateway SDK for typed FinOps metadata.

Inference providers

OpenAIAnthropicGeminiAzure OpenAIAWS BedrockVertex AIOllamaMistralCohereTogetherGroqDeepSeek+24 more

How it works

Three steps from your first request to full FinOps governance.

01

Route through the gateway

Send requests to the CoreValue AI Gateway instead of your provider directly. One API key unlocks 35+ providers with pass-through routing and automatic fallbacks.

02

We log and bill everything

The gateway captures request and response data asynchronously — no latency added to your users. Cost is calculated per call, per user, per team. Metadata goes to Postgres, analytics to ClickHouse.

03

Govern, analyze, optimize

Set budget gates and alerts, track cost by user and team, and export chargeback reports — all from one dashboard.

Simple, transparent pricing

Start free with 1,000 requests a month. Upgrade to Growth for metered usage or contact us for enterprise scale.

Free

For getting started and small projects.

$0/ forever
Start free
  • 1 seat
  • 1 organization
  • 1,000 requests / month
  • Dashboard, requests & cost tracking
  • 14-day data retention
  • Rate limiting (1 alert)

Growth

For production teams monitoring LLM apps.

$30/ seat / mo
Upgrade to Growth
  • 1 organization
  • 3,000 requests included, then starting at $0.0007/req
  • Metered storage starting at $0.002/GB
  • Alerts, caching, budget gates & more
  • ROI scorecards & executive reports
  • 90-day data retention
  • Email support

Enterprise

For organizations with scale & compliance needs.

Contact sales
  • Unlimited organizations
  • Custom requests volume
  • Unlimited alerts & rate limits
  • SSO & RBAC
  • Security & Compliance
  • Custom data retention
  • Priority support & dedicated SLA

Reasonable questions

Everything you need to know before integrating.

Swap your provider SDK's baseURL to point at the CoreValue gateway. Same surface, same types. We record the full request and response on the wire, with no extra round-trip and no sampling by default. Works with the OpenAI SDK, Anthropic SDK, Google GenAI SDK, LangChain, Vercel AI SDK, and more. Or use the @cova/gateway SDK for typed FinOps metadata.

Ship your first request through CoreValue in 2 minutes

No credit card. No infrastructure to manage. Just point your base URL at CoreValue and start governing your AI spend.