getRecentPrioritizationFees — Solana RPC Method Guide
getRecentPrioritizationFees is a native Solana JSON-RPC method that returns a list of minimum prioritization fees observed in the most recent slots (up to 150 blocks). It is the foundation for any dynamic fee estimation system on Solana.
Method Signature
getRecentPrioritizationFees(accountAddresses?: string[]) → Array of {slot: number, prioritizationFee: number}
The optional accountAddresses parameter is critical for accurate localized fee data. When provided, the RPC returns fee samples only from transactions that locked at least one of the specified accounts — giving you program-specific fee floors rather than global network averages.
Understanding the Response
Each item in the response array contains:
- slot: The slot number for this fee sample
- prioritizationFee: The minimum priority fee (in microLamports per CU) observed in that slot for the queried accounts
A prioritizationFee of 0 means transactions in that slot had no priority fee attached — useful for identifying when the network is lightly loaded.
Limitations of the Raw RPC Method
The raw method returns up to 150 slot samples, but interpreting them requires additional processing:
- The data represents minimum observed fees — not average or median fees
- Rapidly changing conditions make the 150-slot window stale within seconds during high volatility
- Single-value extraction (e.g., taking the maximum) can lead to over-paying during brief spikes
For these reasons, the Helius Priority Fee API wraps this method with statistical analysis and returns a single recommended value per priority tier — simplifying integration significantly.
Best Practices for Using getRecentPrioritizationFees
- Always pass the relevant account addresses (program ID + writable accounts) for localized estimates
- Filter out zero-fee slots before calculating your median or percentile
- Use the 50th–75th percentile of non-zero samples for standard production transactions
- Cache results for 2–5 seconds between transaction batches to reduce RPC load
- Cross-reference with global tracker data during unusual network conditions
Related Resources
About The Editor
