Fix streaming doesn't return finish reason (#798)

When streaming the yield that contains the finish can be skipped. This change ensures that yield isn't skipped.
2023-10-19 02:55:56 -04:00 · 2023-10-19 02:55:56 -04:00 · 09a8406c83
commit 09a8406c83
parent 28c2b884e2
1 changed files with 14 additions and 28 deletions
--- a/llama_cpp/llama.py
+++ b/llama_cpp/llama.py
@ -1232,20 +1232,6 @@ class Llama:
                            }
                        ],
                    }
-                    yield {
-                        "id": completion_id,
-                        "object": "text_completion",
-                        "created": created,
-                        "model": model_name,
-                        "choices": [
-                            {
-                                "text": "",
-                                "index": 0,
-                                "logprobs": None,
-                                "finish_reason": finish_reason,
-                            }
-                        ],
-                    }
                    break
                returned_tokens += 1
                yield {