Distribute and run LLMs with a single file.
C++ cross-platformggufllama-cpplocal-ailocal-inferencelocal-llmopen-source-aisingle-file-executablespeech-to-text
Fix JSON parser bug
Justine Tunneyfb4b3e6e454d
diff --git a/llamafile/server/json.cpp b/llamafile/server/json.cppindex 4cd4ba5a..6a78bf24 100644--- a/llamafile/server/json.cpp+++ b/llamafile/server/json.cpp@@ -951,6 +951,8 @@ Json::parse(Json& json, const char*& p, const char* e, int context, int depth)return success;if (status != success)return status;+ if (!key.isString())+ return object_key_must_be_string;status = parse(value, p, e, COLON, depth - 1);if (status == absent_value)return object_missing_value;