How to use GeckoTerminal (and its free API)
The Editor·8 min read·Updated 31 Aug 2026
How to use GeckoTerminal: finding pools, reading the pair page, and pulling OHLCV and new-pool data from the free keyless API without paying for a data feed.
GeckoTerminal is CoinGecko's on-chain product: it indexes decentralised exchange pools across many networks and shows price, volume, liquidity and a chart per pool. As a manual screener it is competent. Its real advantage is that the same data comes out of a free public API with no key, which is why most people arrive for the second reason.
What GeckoTerminal is, and what it is not
It indexes pools, not projects. Every page describes one pair on one DEX on one network, exactly as DexScreener does — and like DexScreener, a token with pools on several DEXs has several pages.
It is a different product from coingecko.com. GeckoTerminal picks up your pool automatically once liquidity exists, with no submission and no approval. A profile on coingecko.com is a curated editorial decision made from a request. A token can be fully visible on the first while having no page at all on the second, and the distinction trips up a lot of launchers — it is worked through in how to get a token listed on CoinGecko and CoinMarketCap.
Against DexScreener, the honest comparison is narrow. DexScreener is faster to reach for and has the wider mindshare; GeckoTerminal has the better free programmatic access. For live manual trading most people will keep using the former, and the field-by-field guide to it is in how to use DexScreener properly. The full tool-by-tool matrix is in the best memecoin screeners and trackers.
Using the interface
Pick the network first. The network selector sits at the top of the site and everything below it is scoped to that choice. Most confused GeckoTerminal sessions are someone looking at the wrong network.
Search by contract address, not by name. The search field accepts a token or pool address. Searching a ticker returns everything that has ever used that string, including deliberate impersonations of whatever is currently getting attention. An address returns exactly the pools for that contract.
Read the pool page from the header down. The header carries the pair, the DEX it trades on, and the price in both USD and the network's base asset. Underneath sit the numbers that matter:
| Field | What it is | What people get wrong |
|---|---|---|
| Liquidity | Total value in the pool, both sides | Read as money spent on the token; roughly half is the token itself |
| FDV | Total supply × price | Read as capital invested. It is a multiplication |
| Market cap | Circulating supply × price | Often identical to FDV on a memecoin with full supply circulating |
| Volume (24h) | Value traded in the window | On a young pool, cheap to manufacture by trading against yourself |
| Transactions / Buyers / Sellers | Counts in the window | Address counts, not people |
| Pool created | Age of the pool | Not the age of the token; a migrated token starts a new pool at zero |
The FDV-versus-market-cap point is the one that costs people money most often, and market cap versus fully diluted valuation for memecoins sets out when the gap is meaningful and when it is noise.
The chart is a pool chart. Candles are built from trades in that pool only. If the token has meaningful volume in a second pool, the chart in front of you is a partial view of its price history, not the whole of it.
New pools and trending. The new-pools view lists recently created pools on the selected network, and there is a trending view alongside it. Both are worth setting a liquidity floor against mentally before reading, for the reasons in how to monitor new token launches in real time.
The free API — the actual reason to use it
This is where GeckoTerminal earns its place. The public API is keyless: no account, no token, no billing setup. That makes it the fastest route from "I want on-chain pool data in a spreadsheet or a script" to having it.
The endpoints you will use most, under the versioned public base path:
| Endpoint shape | What it returns |
|---|---|
/networks | The list of supported networks and their identifiers |
/networks/{network}/dexes | The DEXs indexed on a network |
/networks/{network}/new_pools | Recently created pools on that network |
/networks/new_pools | Recently created pools across networks |
/networks/{network}/trending_pools | Trending pools on that network |
/networks/{network}/pools/{pool_address} | Full data for a single pool |
/networks/{network}/tokens/{token_address} | Token-level data, including its pools |
/networks/{network}/pools/{pool_address}/ohlcv/{timeframe} | Candles for a pool at a given timeframe |
/simple/networks/{network}/token_price/{addresses} | Prices for a batch of token addresses |
The OHLCV endpoint is the one that has no comparable free alternative. Historical candles for an arbitrary DEX pool, at several timeframes, without a key, is not available from most of this category at any price tier that a hobbyist would reach.
Three practical notes before you build anything.
Get the network identifier right. Networks are referenced by GeckoTerminal's own slugs, not by chain ID and not by the name you would use in conversation. Pull /networks once and keep the mapping; guessing the string is the most common first failure.
Send the version header. The public API expects an Accept header carrying the API version alongside the JSON content type. Requests that omit it can behave differently after a version change, which produces the kind of bug that appears weeks later.
Read the current rate limit from the docs, not from an article. The keyless tier has a low fixed per-minute limit, and it is a real constraint: a naive script that polls new pools across several networks on a short interval will hit it. GeckoTerminal publishes the current number on its own API documentation, along with the current base path and version string. Those change; a figure printed here would be wrong before you read it, which is why there is not one.
A working pattern for pulling your own data
- Call
/networksonce and store the identifiers for the networks you care about. - Resolve the pool, not the token. For a specific token, call the token endpoint first and choose the pool with the deepest liquidity. Pulling prices for the wrong pool is the most common silent error in this kind of script.
- Pull OHLCV per pool at the coarsest timeframe that answers your question. Minute candles across many pools will exhaust your rate budget quickly and are usually not what the analysis needed.
- Cache aggressively and back off on errors. Respect the documented limit rather than discovering it. Exponential backoff on a 429 costs three lines.
- Persist what you pull. Historical windows on free endpoints are bounded. If you want a long series, you build it by storing what you collect, starting now.
- Handle the empty case. A brand-new pool can return a valid response with almost no history and a price derived from two trades. Your code should treat that as a data-quality condition, not as a price.
If what you actually want is launch-level rather than pool-level data — which venue a token came from, its launch model, and a per-chain safety report attached — that is a different shape of question, and Meme Central's public data API exposes it. It covers tokens indexed by Meme Central rather than the whole market, which is a genuine limitation of any index including CoinGecko's.
What this doesn't tell you
GeckoTerminal reports pools accurately. It does not assess them. Nothing on a pool page tells you whether the contract can mint further supply, freeze balances or tax sells, or whether the liquidity behind it can be withdrawn — those are contract properties requiring a security tool, and a pool with clean-looking numbers and a hostile contract renders identically to one without.
The free API is generous rather than unlimited. Rate limits, historical depth and endpoint availability are set by CoinGecko and have changed before; a paid CoinGecko tier exists that covers the same on-chain data with higher limits. If your project depends on this data, read the terms on the source rather than assuming the free tier is permanent.
And coverage is not universal. Chains, DEXs and individual long-tail pools can be missing or delayed, and a pool that does not appear has not been judged — it has not been indexed.
Frequently asked questions
Is the GeckoTerminal API free?
The public API is free and keyless: no account and no billing. It carries a low documented per-minute rate limit and bounded historical depth, and CoinGecko sells paid tiers covering the same on-chain data with higher limits. Check the current limits on GeckoTerminal's own API documentation before you build against them.
What is the difference between GeckoTerminal and CoinGecko?
GeckoTerminal is the on-chain DEX product: it indexes liquidity pools automatically, with no submission. coingecko.com is the curated aggregator, where a token profile is created by a listing team from a request. A token can be fully visible on GeckoTerminal while having no coingecko.com page at all.
Is GeckoTerminal better than DexScreener?
They are good at different things. DexScreener is the faster manual screener with the wider coverage of chains and the larger user base. GeckoTerminal has the better free programmatic access, and its OHLCV endpoint has no comparable keyless alternative. Most people use the first to look and the second to pull.
Can I get historical candles for any DEX pool?
For indexed pools, yes, at several timeframes, through the OHLCV endpoint. The depth of history returned is bounded, so if you need a long series the practical approach is to start collecting and storing it now rather than expecting to backfill it later.
The property no data feed can report for you
An API will tell you what is in a pool to the cent. It will not tell you whether that liquidity can be pulled tomorrow, because that depends on who controls the LP tokens. Locking liquidity with Team Finance — built by TrustSwap, which also builds Meme Central — holds them for a fixed term on Ethereum, Robinhood Chain, Polygon, Base and BNB, and where a lock exists it shows as a verified badge on the token's page in the cross-venue launch feed. It settles that one question. It says nothing about the supply a deployer still holds, and no lock makes a thin pool deep.
Nothing here is financial, legal or tax advice. Memecoins are extremely high-risk: most lose most of their value, and the majority of tokens launched never reach a decentralised exchange at all. Never spend money you cannot afford to lose entirely. Meme Central does not recommend any specific token. Data described as Meme Central's own reflects tokens indexed by Meme Central and is not whole-market data.
