Skip to main content
Submits a signed transaction to the cluster for processing. Returns the transaction signature on success. For the best transaction landing rates, use a SWQoS API key and set skipPreflight: true with maxRetries: 0. See the sending transactions guide for more detail.

​
Request

{
  "jsonrpc": "2.0",
  "id": 1,
  "method": "sendTransaction",
  "params": [
    "SIGNED_TX_BASE64",
    { "encoding": "base64", "skipPreflight": false }
  ]
}

​
Parameters

​
transaction
string
required
Signed, serialized transaction encoded as a base-58 or base-64 string.
​
encoding
string
Encoding of the transaction parameter. base58 or base64. Defaults to base58. Use base64 for versioned transactions.
​
skipPreflight
boolean
Skip the preflight simulation step. Defaults to false. Set to true when using SWQoS keys for lower latency.
​
preflightCommitment
string
Commitment level used for preflight simulation. Defaults to finalized.
​
maxRetries
number
Maximum number of times the RPC node should retry sending the transaction. Set to 0 when using SWQoS to avoid duplicate sends.

​
Response

{
  "jsonrpc": "2.0",
  "id": 1,
  "result": "5yGsRzg4MCzTBq5mHJSWAtCZEXh6ZBF7CBrHp4AHiuuTKe5j..."
}
​
result
string
The transaction signature as a base-58 encoded string. Use this with getTransaction to confirm the transaction.
A successful response means the transaction was accepted by the node, not that it was confirmed on-chain. Use getSignaturesForAddress or a WebSocket subscription to track confirmation.

​
Rate limits

sendTransaction has a separate, lower rate limit from other RPC methods.
PlansendTransaction/s
Free1
Dev5
Pro20
Ultra50
Shared Metal200

​
Example

curl "https://rpc.solami.fast/sol?api_key=YOUR_API_KEY" \
  -X POST \
  -H "Content-Type: application/json" \
  -d '{
    "jsonrpc": "2.0",
    "id": 1,
    "method": "sendTransaction",
    "params": [
      "SIGNED_TX_BASE64",
      { "encoding": "base64", "skipPreflight": false }
    ]
  }'
When using a SWQoS key, set skipPreflight: true and maxRetries: 0 for the lowest possible latency.