In my understanding, smart contracts on Cardano are always provided to the blockchain via a transaction and there is no code that is stored at the blockchain, but just a hash (the public key address). Ethereum seems to store them instead: see What is the maximum number of contracts Cardano can store?
Is this correct?
What are the other main technical differences?
Update
Also beyond that Cardano uses the eUTxO-based model and Ethereum uses an account-based model.
Feel free to improve! :)
| technical aspect | Cardano | Ethereum | notes |
|---|---|---|---|
| model | eUTxO-based | account-based | see cardano lecture, see question, ... |
| smart contract code location | not stored on-chain, but shipped with transaction | stored on-chain ? | |
| smart contract maximum count | unlimited | limited ? | see question |
| access third party contract e.g. an oracle | import and use lib (required or just helpful?), , compile your own contract, ship third party validator script with transaction, ... |
call it via an address ? | see question |
| ... | ... | ... | ... |
I'd like to complement @JohannesSirsch answer.
One of the most important aspects of any network, when working with smart contracts, is the ability of the script to see the "event horizon":
This video lecture in the Plutus Pioneers Program explains it https://youtu.be/igV7kMXcdpw
