Platform Overview
Understand how CoreValue solves the core challenges of building production LLM applications
Now that your requests are flowing through CoreValue, let's explore what you can do with the platform.
What is CoreValue?
CoreValue is an AI Gateway and observability platform for production LLM applications. It provides a single unified API to route requests across multiple LLM providers, with built-in fallbacks, caching, rate limiting, and real cost tracking — plus full observability through the dashboard.
In short: route intelligently, monitor everything, never go down.
The Problems We Solve
Reliability Issues
Provider outages break your application. No visibility when requests fail. Manual fallback logic is complex and error-prone.
Debugging Complexity
LLM responses are non-deterministic. Multi-step AI workflows are hard to trace. Errors are difficult to reproduce.
Cost Uncertainty
Unpredictable spending across providers. No understanding of unit economics. Difficult to optimize without breaking functionality.
Prompt Management Pain
Every prompt change requires a deployment. No version control for prompts. Can't iterate quickly based on user feedback.
How It Works
CoreValue works in two ways: use our AI Gateway with pass-through billing (easiest), or bring your own API keys for observability-only mode.
Option 1: AI Gateway (Recommended)
Access multiple LLM models through a single unified API with zero markup:
- Add Credits - Top up your CoreValue account (0% markup)
- Single Integration - Point your OpenAI SDK to our gateway URL
- Use Any Model - Switch between providers by just changing the model name
- Automatic Observability - Every request is logged with costs, latency, and errors tracked
No need to sign up for OpenAI, Anthropic, Google, or any other provider. We manage the API keys and you get complete observability built in.
Gateway Features
The AI Gateway (https://gateway.corevalue.dev/v1) is an OpenAI-compatible API that routes your requests to the best provider. Here's what's built in:
| Feature | How it works |
|---|---|
| Provider Routing | Use provider-prefixed model names (e.g. openai/gpt-4o, anthropic/claude-3.5-sonnet) to route to any supported provider. |
| Fallback Chains | Set Cova-Fallbacks to a JSON array of model names (max 5). If the primary model fails, the gateway tries the next one automatically. |
| Caching | Set Cova-Cache-Enabled: true to cache responses. Control TTL, bucket size, seed, and ignore keys via Cova-Cache-* headers. |
| Rate Limiting | Set Cova-RateLimit-Policy to limit requests by count, cost, or per-user/per-property segments. |
| Retries | The gateway automatically retries transient failures (429, 500, 502, 503, 504, 529) with exponential backoff. Configured server-side; observe X-Cova-Retry-Count on the response. |
| Real Cost Tracking | Every response includes Cova-Request-Cost with the actual cost in USD — not an estimate. |
| Cross-Provider Translation | Send OpenAI-format requests to Anthropic, Google, or vice versa. The gateway translates request and response formats automatically. |
| Custom Properties | Add Cova-Property-{Name} headers to tag requests with custom metadata for filtering and analytics. |
| User Tracking | Set Cova-User-Id to track per-user costs, usage, and activity. |
| Prompt Tracking | Set Cova-Prompt-Id to correlate requests with prompt versions in the dashboard. |
| Feedback | POST /v1/feedback to attach user feedback (thumbs up/down, comments) to any request. |
| Meta Endpoint | GET /gateway/v1/meta/:requestId to retrieve cost, provider, model, and status metadata for any request. |
| Models Endpoint | GET /v1/models to list all available models in OpenAI-compatible format. |
See the Gateway Overview for details on each feature.
Dashboard Features
Every request through the gateway is automatically logged. The CoreValue dashboard gives you:
Sessions
Group related requests into sessions to trace multi-step AI workflows and agent conversations.
Datasets
Create datasets from request logs for fine-tuning, evaluation, or batch analysis.
Experiments
Run A/B experiments comparing different prompts, models, or parameters side by side.
Evaluators
Score and evaluate responses with custom evaluators or user feedback.
Prompts
Manage prompt templates with versioning and environments. Deploy prompt changes without code deploys.
Reports
Generate reports on usage, costs, and performance across your organization.
Webhooks
Receive real-time webhook notifications for request events.
User Metrics
Track per-user costs, usage patterns, and activity with Cova-User-Id.
Our Principles
Best Price Always We fight for every penny. 0% markup on credits means you pay exactly what providers charge. No hidden fees, no games.
Invisible Performance Your app shouldn't slow down for observability. The Hono gateway on Cloud Run keeps us fast. Always.
Always Online Your app stays up, period. Providers fail, we fallback. Rate limits hit, we load balance. We don't go down.
Never Be Surprised
No shock bills. No mystery spikes. See every cost as it happens with Cova-Request-Cost on every response.
Find Anything Every request, searchable. Every error, findable. That needle in the haystack? We'll help you find it.
Built for Your Worst Day When production breaks and everyone's panicking, we're rock solid. Built for when you need us most.
Real Scenarios
Start Exploring
AI Gateway
Learn how the gateway routes requests, handles fallbacks, and tracks real costs.
SDK Quick Start
Get started with the @cova/gateway SDK in TypeScript or Python.
Sessions
Debug complex AI agents and multi-step workflows with session tracking.
Prompt Management
Deploy prompts without code changes using versioned prompt templates.
Questions? Contact support@corevalue.dev with any questions or feedback.