Skip to main content
This page collects the most useful answers for teams evaluating, integrating, or operating Uniblock in production.
Looking for a full product introduction? Start with Uniblock Overview.

Getting Started

First API call, authentication basics, and quickstart path.

Routing and Reliability

Auto-routing, backup behavior, and provider failover.

Errors and Debugging

429/4xx guidance, JSON-RPC behavior, and debug flow.

Support and Coverage

APIs, chain/provider coverage, MCP, and support.

​
At a glance

​
What is Uniblock?

Uniblock is a blockchain infrastructure layer that gives you one unified API across many chains and providers. Instead of integrating directly with providers like Alchemy, QuickNode, Infura, Moralis, or Covalent one by one, you send requests through Uniblock and get:
  • One API key
  • Auto-routing across providers
  • Backup failover when providers have issues
  • Normalized response formats
  • Access to many supported chains through the same interface
If you are just getting started, use the Unified API Quickstart.

​
Why use Uniblock instead of integrating with a single provider directly?

Use Uniblock when you want to reduce provider management overhead and improve reliability. A direct provider integration can be simpler if you only need one chain and one provider, but Uniblock is usually more valuable when you need:
  • Multi-chain coverage
  • Automatic failover
  • Consistent schemas across providers
  • Centralized analytics and logs
  • One integration instead of managing multiple provider APIs and keys

​
How do I make my first API call?

  1. Create a project in the dashboard.
  2. Copy your Uniblock API key.
  3. Send the key in the request header.
  4. Call a Unified API endpoint.
Example:
For a step-by-step walkthrough, see Unified API Quickstart.

​
How do I authenticate requests?

Include your Uniblock API key in the request header as x-api-key. Every request to the Unified API should include a valid project key.

​
Can I pass my API key in the URL or as a query parameter?

No. Uniblock HTTP endpoints (Unified API, the Unified JSON-RPC endpoint at https://api.uniblock.dev/uni/v1/json-rpc, and Direct API endpoints) accept the key only in the X-API-Key header. There is no URL format that embeds the key in the path or query string. WebSocket connections are the exception. Most use the X-API-KEY header, but some provider-specific endpoints (such as Codex) accept apiKey as a query parameter. See WebSocket authentication. For full details and examples in cURL, JavaScript, and Python, see Authentication.

​
Is auto-routing enabled by default?

Yes. Auto-routing is enabled by default for all Uniblock API keys. You do not need to manually choose a provider for standard usage. Uniblock evaluates providers for reliability, latency, and cost, routes to the best available option, retries failures, and fails over to backups when needed. See Auto-routing for the full flow.

​
Do backup providers prevent all failures?

No. Backup providers help when an upstream provider is unavailable, returns errors, or hits its own provider-specific limit. They do not override Uniblock plan-level limits. If Uniblock returns a 429, that means your project has reached the request limit included with your current plan. That is different from a provider-specific failure inside the routing layer. See:

​
What does a 429 mean?

A 429 means your project has reached the request limit available under your current Uniblock plan. It does not mean:
  • The API is down
  • A provider outage occurred
  • Backup routing failed
In most cases, you should reduce request frequency, retry with backoff, or move to a plan with more throughput.

​
Does Uniblock normalize provider responses?

Yes. Uniblock transforms provider-specific responses into a unified format so your application does not need provider-specific parsing logic for common Unified API endpoints. This is one of the main reasons to use Uniblock for multi-provider and multi-chain applications.

​
Can I bring my own provider keys?

Yes. Uniblock supports custom fallback lists with external provider keys, which lets you consume your own provider credits first and keep Uniblock auto-routing as a final fallback. This is useful if you already have relationships or spend commitments with specific providers. See Backup Providers.

​
What is data consensus, and when should I use it?

Data consensus queries multiple providers and compares results to improve confidence in the returned data. It is most useful for critical reads where correctness matters more than the extra latency and cost. Use it selectively for things like:
  • Critical balance checks
  • Pricing-sensitive workflows
  • Auditing or verification flows

​
Why did I get a 403?

Common reasons include:
  • Your API key or project credentials are missing or invalid
  • The API key has been disabled
  • Your project does not have permission for the requested endpoint
  • Your project does not have access to the requested blockchain network
If the response includes a Uniblock error code, check Error Codes & Rate Limits first and then verify your project configuration in the dashboard.
Fast triage for 403: verify API key validity, endpoint access, and chain permissions in the dashboard before retrying.

​
Why did I get a 400?

A 400 usually means the request itself is invalid. Common examples include:
  • Invalid query or body parameters
  • Invalid blockchain address format
  • Invalid transaction hash format
  • Unsupported JSON-RPC method
  • Pagination or cursor errors
These are usually not retryable until you fix the request.

​
Why did I get a 422?

A 422 means the upstream provider returned a successful HTTP response, but Uniblock could not safely parse, transform, or validate the content. This usually points to a provider response problem rather than a malformed request from your application. If it persists:
  1. Retry after a short delay.
  2. Check API logs and request details.
  3. Contact support with the request ID if the issue continues.
Persistent UR-422-* errors usually indicate an upstream response issue. Include request ID and endpoint when contacting support.

​
Why do I see “gas required exceeds” on eth_call or eth_estimateGas?

Public JSON-RPC providers apply a gas cap to eth_call and eth_estimateGas to protect shared infrastructure. The most common cap is 16777216 (2^24, roughly 16.7 million), which produces errors like:
This cap is enforced by the upstream provider that Uniblock routed the request to, not by Uniblock itself. It applies to simulation and gas-estimation calls, and to large contract deployments that would exceed the cap during estimation. To work around it:
  1. Bring your own provider key. Add an API key from a provider whose plan allows a higher gas cap and route through it via Add Your Own API Key. Requests then consume your provider’s quota with its configured limits.
  2. Use a direct provider integration. For provider-specific behavior that the shared RPC does not expose, use a Direct Provider endpoint tied to a provider account you control.
  3. Set an explicit gas value. If your client is calling eth_estimateGas before a transaction, pass a known-good gas value in the transaction to skip estimation.
  4. Contact support. If none of the above work for a deployment or simulation you need to run, reach out with the request ID and target chain so the team can advise on the right routing.

​
Why do some JSON-RPC errors return HTTP 200?

JSON-RPC errors can be returned inside a valid JSON-RPC response body while the HTTP layer still returns 200. In those cases, the transport succeeded, but the JSON-RPC request itself failed. That is why you may see things like:
  • Parse error
  • Invalid request
  • Batch size exceeded
with an HTTP 200 plus a JSON-RPC error payload.

​
How do I debug failed requests?

Start with these steps:
  1. Check the HTTP status code.
  2. Check the Uniblock error code if present.
  3. Review the request parameters and headers.
  4. Inspect API logs in the dashboard.
  5. Look at provider attempts and failures if failover occurred.
The most useful docs are:

​
Debug checklist

1

Classify by status code

Determine whether the failure is auth/permissions (403), request-shape (400), rate-limit (429), or response-processing (422).
2

Inspect request + context

Confirm headers, params, chain ID, and endpoint path. Keep the request ID for escalation.
3

Check logs and failover behavior

Use dashboard API logs to inspect latency, provider attempts, and failover paths.
4

Escalate with evidence

Share request ID, endpoint, time window, and sample payload with support.

​
Where can I see which provider handled my request?

Use the API logs in the dashboard. Logs help you inspect request status, endpoint, response time, and provider activity, including failover attempts where applicable.

​
Is there a free plan?

No. The entry tier is Startup, which includes 40 million Compute Units (CU) per month and 1,000 CU/s of throughput. All paid tiers include the Unified API, JSON-RPC, and Direct API endpoints. See Pricing Plans for the full comparison.

​
How do I map monthly requests to a plan?

Plans are metered in Compute Units (CU), not raw request counts. Different endpoints and JSON-RPC methods weigh different amounts of CU, so the same number of requests can consume very different amounts of CU depending on which endpoints you call. To pick a plan:
  1. Estimate your monthly traffic in the pricing estimator, which converts expected traffic into CU.
  2. Match the estimated CU to a tier on Pricing Plans.
  3. For a per-endpoint breakdown of CU weights, see the Compute Units overview.

​
Which APIs does Uniblock support?

Uniblock provides unified access across multiple API categories, including:
  • Token APIs
  • NFT APIs
  • Transaction APIs
  • Market data APIs
  • Scans APIs
  • JSON-RPC
The exact endpoint list is available in the API reference and explorer.

​
Which chains and providers are supported?

Uniblock supports many blockchains and integrates with many blockchain data providers. The docs describe support for 300+ blockchains and 55+ providers. For chain-by-chain details, check the supported chains and API reference sections in the docs.

​
Does Uniblock support MCP?

Yes. Uniblock provides an MCP server that can be connected to compatible AI clients. This exposes Uniblock tools and Unified API endpoints through MCP. See MCP Setup.

​
What are Uniblock’s official domains?

Uniblock operates across multiple domains for marketing and outreach. Use the list below to verify you’re interacting with an official Uniblock property.

​
When should I contact support?

Contact support when:
  • You continue seeing the same UR-422-* response-processing issue
  • Your project appears misconfigured but the dashboard does not explain why
  • Your API key was disabled unexpectedly
  • You need higher limits or custom access
  • You have a request ID tied to a persistent internal or unclassified error

Ready to get started with Uniblock?

Create your free account and start building with unified blockchain infrastructure across 300+ chains.

View API reference

Explore all available endpoints and interactive playgrounds