Support defaulting to infinity or -1 for completions

llama_cpp/llama.py

@@ -1917,7 +1917,7 @@ def create_completion(
         completion_or_chunks = self._create_completion(
             prompt=prompt,
             suffix=suffix,
-            max_tokens=max_tokens,
+            max_tokens=-1 if max_tokens is None else max_tokens,
             temperature=temperature,
             top_p=top_p,
             min_p=min_p,
@@ -1951,7 +1951,7 @@ def __call__(
         self,
         prompt: str,
         suffix: Optional[str] = None,
-        max_tokens: int = 128,
+        max_tokens: Optional[int] = 16,
         temperature: float = 0.8,
         top_p: float = 0.95,
         min_p: float = 0.05,

llama_cpp/server/errors.py

@@ -72,7 +72,7 @@ def context_length_exceeded(
         return 400, ErrorResponse(
             message=message.format(
                 context_window,
-                completion_tokens + prompt_tokens,
+                (completion_tokens or 0) + prompt_tokens,
                 prompt_tokens,
                 completion_tokens,
             ), # type: ignore

llama_cpp/server/types.py

@@ -110,7 +110,9 @@ class CreateCompletionRequest(BaseModel):
         default=None,
         description="A suffix to append to the generated text. If None, no suffix is appended. Useful for chatbots.",
     )
-    max_tokens: int = max_tokens_field
+    max_tokens: Optional[int] =  Field(
+        default=16, ge=0, description="The maximum number of tokens to generate."
+    )
     temperature: float = temperature_field
     top_p: float = top_p_field
     min_p: float = min_p_field