From a928c62b501572071cee2768cea2e65d4a1d50eb Mon Sep 17 00:00:00 2001
From: Sloane <sloane.simmons@toyotaconnected.com>
Date: Thu, 1 Aug 2024 15:50:27 -0500
Subject: [PATCH] fix llama_grammar_accept_token arg order

Old was:
llama_grammar_accept_token(ctx, grammar, token)

Now this is:
llama_grammar_accept_token(grammar, ctx, token)
---
 llama_cpp/_internals.py | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/llama_cpp/_internals.py b/llama_cpp/_internals.py
index dcd4e17ff..357023e3c 100644
--- a/llama_cpp/_internals.py
+++ b/llama_cpp/_internals.py
@@ -511,7 +511,7 @@ def sample_token(self, candidates: "_LlamaTokenDataArray") -> int:
     def grammar_accept_token(self, grammar: LlamaGrammar, token: int):
         assert self.ctx is not None
         assert grammar.grammar is not None
-        llama_cpp.llama_grammar_accept_token(self.ctx, grammar.grammar, token)
+        llama_cpp.llama_grammar_accept_token(grammar.grammar, self.ctx, token)
 
     def reset_timings(self):
         assert self.ctx is not None