diff --git a/parser/src/tokenparser.rs b/parser/src/tokenparser.rs
index b489241..d5f37d5 100644
--- a/parser/src/tokenparser.rs
+++ b/parser/src/tokenparser.rs
@@ -254,12 +254,32 @@ impl TokenParser {
         Ok(())
     }
 
+    pub fn validate_token(&mut self, token: TokenId) -> Result<bool> {
+        self.check_initialized("validate_tokens_raw")?;
+        let bytes = self.tok_trie().decode_raw(&[token]);
+        let n_valid = self.parser.validate_bytes(&bytes);
+        assert!(n_valid <= bytes.len());
+        Ok(n_valid == bytes.len())
+    }
+
     /// Returns how many of the passed tokens can be accepted by the parser.
     /// It does not tokenize forced bytes, so will accept non-canonical tokenizations.
     /// If called with more than one token, it may ignore max_tokens constraints.
     pub fn validate_tokens_raw(&mut self, tokens: &[TokenId]) -> Result<usize> {
         self.check_initialized("validate_tokens_raw")?;
 
+        if tokens.is_empty() {
+            return Ok(0);
+        }
+
+        if tokens.len() == 1 {
+            return if self.validate_token(tokens[0])? {
+                Ok(1)
+            } else {
+                Ok(0)
+            };
+        }
+
         let bytes = self.tok_trie().decode_raw(tokens);
         let n_valid = self.parser.validate_bytes(&bytes);
         assert!(n_valid <= bytes.len());