diff --git a/client/src/rpc_filter.rs b/client/src/rpc_filter.rs
index 416742d591..500f74f9c0 100644
--- a/client/src/rpc_filter.rs
+++ b/client/src/rpc_filter.rs
@@ -16,10 +16,15 @@ impl RpcFilterType {
                 match encoding {
                     MemcmpEncoding::Binary => {
                         let MemcmpEncodedBytes::Binary(bytes) = &compare.bytes;
-                        bs58::decode(&bytes)
-                            .into_vec()
-                            .map(|_| ())
-                            .map_err(|e| e.into())
+
+                        if bytes.len() > 128 {
+                            Err(RpcFilterError::Base58DataTooLarge)
+                        } else {
+                            bs58::decode(&bytes)
+                                .into_vec()
+                                .map(|_| ())
+                                .map_err(|e| e.into())
+                        }
                     }
                 }
             }
@@ -27,10 +32,12 @@ impl RpcFilterType {
     }
 }
 
-#[derive(Error, Debug)]
+#[derive(Error, PartialEq, Debug)]
 pub enum RpcFilterError {
     #[error("bs58 decode error")]
     DecodeError(#[from] bs58::decode::Error),
+    #[error("encoded binary (base 58) data should be less than 129 bytes")]
+    Base58DataTooLarge,
 }
 
 #[derive(Debug, Clone, PartialEq, Serialize, Deserialize)]
@@ -140,4 +147,36 @@ mod tests {
         }
         .bytes_match(&data));
     }
+
+    #[test]
+    fn test_verify_memcmp() {
+        let base58_bytes = "\
+            1111111111111111111111111111111111111111111111111111111111111111\
+            1111111111111111111111111111111111111111111111111111111111111111";
+        assert_eq!(base58_bytes.len(), 128);
+        assert_eq!(
+            RpcFilterType::Memcmp(Memcmp {
+                offset: 0,
+                bytes: MemcmpEncodedBytes::Binary(base58_bytes.to_string()),
+                encoding: None,
+            })
+            .verify(),
+            Ok(())
+        );
+
+        let base58_bytes = "\
+            1111111111111111111111111111111111111111111111111111111111111111\
+            1111111111111111111111111111111111111111111111111111111111111111\
+            1";
+        assert_eq!(base58_bytes.len(), 129);
+        assert_eq!(
+            RpcFilterType::Memcmp(Memcmp {
+                offset: 0,
+                bytes: MemcmpEncodedBytes::Binary(base58_bytes.to_string()),
+                encoding: None,
+            })
+            .verify(),
+            Err(RpcFilterError::Base58DataTooLarge)
+        );
+    }
 }
diff --git a/docs/src/developing/clients/jsonrpc-api.md b/docs/src/developing/clients/jsonrpc-api.md
index 093e9469c5..1fe1744816 100644
--- a/docs/src/developing/clients/jsonrpc-api.md
+++ b/docs/src/developing/clients/jsonrpc-api.md
@@ -207,7 +207,7 @@ Returns all information associated with the account of provided Pubkey
 fields:
   - (optional) [Commitment](jsonrpc-api.md#configuring-state-commitment)
   - `encoding: <string>` - encoding for Account data, either "base58" (*slow*), "base64", "base64+zstd", or "jsonParsed".
-    "base58" is limited to Account data of less than 128 bytes.
+    "base58" is limited to Account data of less than 129 bytes.
     "base64" will return base64 encoded data for Account data of any size.
     "base64+zstd" compresses the Account data using [Zstandard](https://facebook.github.io/zstd/) and base64-encodes the result.
     "jsonParsed" encoding attempts to use program-specific state parsers to return more human-readable and explicit account state data. If "jsonParsed" is requested but a parser cannot be found, the field falls back to "base64" encoding, detectable when the `data` field is type `<string>`.
@@ -1806,7 +1806,7 @@ Returns all accounts owned by the provided program Pubkey
 ##### Filters:
 - `memcmp: <object>` - compares a provided series of bytes with program account data at a particular offset. Fields:
   - `offset: <usize>` - offset into program account data to start comparison
-  - `bytes: <string>` - data to match, as base-58 encoded string
+  - `bytes: <string>` - data to match, as base-58 encoded string and limited to less than 129 bytes
 
 - `dataSize: <u64>` - compares the program account data length with the provided data size