Add batch inference support (WIP)

docs/api-reference.md

@@ -28,16 +28,6 @@ High-level Python bindings for llama.cpp.
             - token_eos
         show_root_heading: true
 
-::: llama_cpp.LlamaGrammar
-    options:
-        members:
-            - from_string
-            - from_json_schema
-
-::: llama_cpp.LlamaCache
-    options:
-        show_root_heading: true
-
 ::: llama_cpp.LlamaState
     options:
         show_root_heading: true
@@ -58,6 +48,13 @@ High-level Python bindings for llama.cpp.
     options:
         show_root_heading: true
 
+::: llama_cpp.LlamaGrammar
+    options:
+        members:
+            - from_string
+            - from_json_schema
+
+
 ## Low Level API
 
 Low-level Python bindings for llama.cpp using Python's ctypes library.

llama_cpp/_internals.py

@@ -767,4 +767,4 @@ def sample(
     def accept(self, ctx_main: _LlamaContext, id: int, apply_grammar: bool):
         if apply_grammar and self.grammar is not None:
             ctx_main.grammar_accept_token(self.grammar, id)
-        self.prev.append(id)
+        self.prev.append(id)