Use Kaspalytics for on-chain history and supply structure, an explorer for a single transaction or address, and a market site for price. Kaspalytics is the deepest free chart set on Kaspa, covering supply aging, exchange holdings, address buckets and covenant activity. For one specific transaction, address or block, use explorer.kaspa.org or the open REST API behind it. For KRC-20 tokens use kas.fyi. For price, volume and exchange data, any market site will do. Nobody is wrong when the numbers differ, they are usually answering different questions.
| Source | Best for | Refresh | Access |
|---|---|---|---|
| Kaspalytics | on-chain history and supply structure: hash rate, difficulty, active addresses, fees, supply aging and HODL waves, exchange holdings, address balance buckets, script types and covenant activity | network and fee views in seconds, supply and address aging about once a day around 00:00 UTC | free, no account |
| kaspa.stream | watching the chain move in real time: blocks and transactions as they land, top addresses, miner and node metrics | live | free, no account |
| explorer.kaspa.org | one specific transaction, address or block, with the same explorer available for the testnets | live | free, no account |
| api.kaspa.org | pulling that same data yourself in code, documented as an OpenAPI page you can click through | live | free, no key |
| kas.fyi | KRC-20 tokens, with a page per token | live | site free, the developer platform at docs.kas.fyi uses API keys and paid plans |
| KasLens | the market side: price across exchanges, volume, order book heatmap, a KRC-20 dashboard | live | free, no account |
| KASLIVE | a live monitoring terminal, including large address movement, a mining calculator and an attack cost estimate | live | free, no account |
| DeFiLlama | money inside the layer 2s, meaning TVL and volume for Kasplex and Igra and the apps on them | live | free, no account |
| Kaspa Pulse | one weekly reading of the numbers that carry weight, with the definition printed next to each one | every Monday | free, no account |
Two things are worth saying plainly. First, almost all of this is free and made by people in this community for nothing, which is not normal for a chain this size. Second, none of these tools is trying to mislead anyone. When two of them disagree, the cause is nearly always a definition, not an error.
This is the one that catches people out most often, so it is worth walking through with the real numbers rather than in the abstract. In August 2026 a live counter on one site was showing somewhere around 11 transactions per second while our own dashboard was publishing 2.34 for the same week. Both figures are defensible. They count different things.
We take the accepted transactions for a full day, standard transactions plus coinbase transactions, and divide by the 86,400 seconds in that day. Accepted matters: Kaspa produces around ten blocks a second in parallel, so the same transaction can appear in more than one block before the DAG settles which of them counts. Counting accepted transactions counts each one once, after that has been decided. Dividing by a whole day means we publish an average across quiet nights and busy afternoons alike, not a moment.
A real-time counter is built to show flow, so it counts what is streaming past right now, which includes the coinbase transaction that every one of those ten blocks a second carries with it. We cannot see anyone else's code and we do not claim to know exactly how another site counts. What we can say is that a per-second live reading and a per-day accepted average are different measurements, and that the gap between them on Kaspa is structural, not a bug in either.
We are a source on this list, so the same standard applies to us. The weekly figures on the dashboard come from Kaspalytics for supply aging, exchange holdings, active addresses and accepted transaction counts, from api.kaspa.org where we count something ourselves transaction by transaction, and from DeFiLlama for layer 2 value. Hash rate and difficulty are cross-checked against more than one source before they go out.
Where we cannot verify a number, we print a dash instead of a guess. That is why the dashboard occasionally shows blanks while other sites show a figure for the same tile. A dash means we could not stand behind it that week, and we would rather be visibly incomplete than quietly wrong. When we do make a claim about what a wallet did, we counted every transaction ourselves and we report the movement, never the motive.