Query Solana with SQL: CDP SQL API public beta

TL;DR: CDP SQL API is now available on Solana Mainnet in public beta. You can query SPL Token and Token-2022 activity with familiar SQL. No indexer, pipeline, or archival infrastructure to run. Query two tables at launch, solana.instructions and solana.transfers, via the SQL Playground, the REST API, or our CoinbaSeQL dialect. It's the same SQL API Base builders already use, extended to Solana, with more programs and tables on the way.

Speed, scale, and the data that follows

Solana is fast, cheap, and scalable, and a lot of developers are choosing to build there. Sub-second finality, fees measured in fractions of a cent, and parallel execution through Sealevel support a wide variety of applications: real-time consumer apps, trading systems, stablecoin payment rails, games with live economies, and high-volume token activity.

That activity shows up in the data. A chain settling this much volume produces a proportionally rich record of it — instruction-level detail on every token movement, across a state model where program data lives in many purpose-built accounts rather than one monolithic contract. It's a precise, expressive design, and it's part of how Solana achieves the throughput it does.

It also means Solana teams have become excellent data engineers. There's sophisticated indexing work happening across this ecosystem: custom Geyser consumers, streaming pipelines, archival strategies, IDL and discriminator management, warehouses tuned for high-cardinality wallet data. That work is impressive.

It's also work that most product teams would rather not have to do twice.

What's launching

Coinbase Developer Platform's SQL API — already used by Base builders to query onchain data with custom SQL — is now available on Solana Mainnet in public beta. The idea is simple: for a large class of questions, you shouldn't need a pipeline at all. You should be able to write a query.

Token coverage at launch: SPL Token and Token-2022 (Token Extensions) — the fungible-token surface most applications reach for first. Additional programs and tables are on the roadmap.

You can reach it three ways, the same as on Base:

• SQL Playground in CDP Portal, no keys required to explore

• REST API for production

• CoinbaSeQL, our SQL dialect for querying indexed onchain data with high freshness

Where this fits in your stack

SQL API isn't a replacement for the infrastructure Solana teams have built. If you're running a trading system that needs microsecond state access, or a custom indexer tuned to your own program's IDL, keep it — a hosted SQL endpoint isn't going to beat a purpose-built pipeline for those workloads.

What SQL API does is take the long tail off your plate. Most teams maintain indexing infrastructure not because indexing is their product, but because a handful of recurring questions demanded it:

• Prototyping and exploration. You want to know whether an idea is worth building before you commit a sprint to the pipeline that would support it. Open the Playground, write a query, get an answer in the browser.

• Product surfaces that aren't hot paths. Portfolio history, transaction detail views, merchant reconciliation, support tooling, internal dashboards. These need correct data more than they need microsecond latency, and they're rarely worth their own infrastructure.

• Historical lookups without archival overhead. Standard RPC keeps a short window by design — it's optimized for the validator's job, not yours. SQL API lets you ask historical questions without standing up and maintaining a separate archival tier for them.

• Multi-chain consistency. If you're already querying Base through SQL API, Solana uses the same shape: network-prefixed tables, SQL in, rows out. One access pattern instead of a bespoke integration per chain.

• Agents and automation. Giving an agent a SQL endpoint against live token activity is considerably more robust than having it maintain its own RPC scrapers.

What you can build with the beta tables

• Payments and commerce: Monitor USDC or any SPL / Token-2022 mint flowing into merchant wallets, and reconcile by time window

• Portfolio and treasury: Reconstruct transfer history across a set of wallets

• Analytics: Rank mints by transfer count or volume, and detect activity spikes without operating your own columnar store

• Ops and support: Pull instruction-level context when a user reports an unexpected token action

• Accounting and reporting: Export token movement over a defined period in a format your finance team can actually use

What's in scope for beta

Shipping solana.instructions and solana.transfers for SPL Token and Token-2022 first means real workloads can validate the product quickly, and we can expand coverage based on concrete demand.

So, to be explicit about today's boundaries:

• Not the full instruction set yet — this is the token transfers  first

• SPL Token and Token-2022 programs at launch, with more programs to follow

• Three months of historical data at this time

• Public beta: coverage, limits, and schema may evolve

If there's a program or table you need next, tell us in #onchain-data on the CDP Discord. That's the channel that shapes the roadmap.

Getting started

1. Open the SQL Playground

2. Query solana.transfers or solana.instructions

3. For production, create a CDP API key and call the Run SQL Query endpoint

4. See the SQL API overview for schema, limits, and best practices

Pricing is pay-as-you-go with a free monthly query allowance — current rates are in the docs.

Bottom line

Solana builders have never been short on capability, and the ecosystem has built serious tooling to handle data at scale.

We'd just like fewer teams to have to build it from scratch. SQL API on Solana is one query away from real token data — no indexer required — and it's live in public beta today.