run-llama/llama_index · error · ValueError

No existing {__name__} found at {persist_path}, skipping loa

Error message

No existing {__name__} found at {persist_path}, skipping load.

What it means

Raised by `SimpleVectorStore.from_persist_path()` when the given persist file does not exist on the provided (or default local) filesystem. The classmethod checks `fs.exists(persist_path)` before opening, and the error message interpolates the module-level `__name__` plus the path. Note the message says 'skipping load' but the code actually raises — the intent is that the caller must create/persist the store first or fix the path.

Source

Thrown at llama-index-core/llama_index/core/vector_stores/simple.py:338

        fs: Optional[fsspec.AbstractFileSystem] = None,
    ) -> None:
        """Persist the SimpleVectorStore to a directory."""
        fs = fs or self._fs
        dirpath = os.path.dirname(persist_path)
        if not fs.exists(dirpath):
            fs.makedirs(dirpath)

        with fs.open(persist_path, "w", encoding="utf-8") as f:
            json.dump(self.data.to_dict(), f)

    @classmethod
    def from_persist_path(
        cls, persist_path: str, fs: Optional[fsspec.AbstractFileSystem] = None
    ) -> "SimpleVectorStore":
        """Create a SimpleKVStore from a persist directory."""
        fs = fs or fsspec.filesystem("file")
        if not fs.exists(persist_path):
            raise ValueError(
                f"No existing {__name__} found at {persist_path}, skipping load."
            )

        logger.debug(f"Loading {__name__} from {persist_path}.")
        with fs.open(persist_path, "rb") as f:
            data_dict = json.load(f)
            data = SimpleVectorStoreData.from_dict(data_dict)
        return cls(data)

    @classmethod
    def from_dict(cls, data: Dict[str, Any], **kwargs: Any) -> "SimpleVectorStore":
        save_data = SimpleVectorStoreData.from_dict(data)
        return cls(save_data)

    def to_dict(self, **kwargs: Any) -> Dict[str, Any]:
        return self.data.to_dict()

View on GitHub (pinned to afd0fef371)

Solutions

1. Run the ingestion/persist step first so the file exists (`index.storage_context.persist()` or `store.persist(path)`).
2. Verify the path: use an absolute path or `os.path.abspath`, and confirm the file exists with `os.path.exists` / `fs.exists`.
3. If a missing store is expected (fresh deploy), branch on existence and build a new store instead of loading.
4. For remote filesystems, double-check the fsspec path format (e.g. bucket/key spelled per the fs's convention).

Example fix

# before
store = SimpleVectorStore.from_persist_path("storage/vector_store.json")  # raises on first run

# after
import os
if os.path.exists(p):
    store = SimpleVectorStore.from_persist_path(p)
else:
    store = SimpleVectorStore()  # build fresh; persist after ingestion

import os

def load_store_or_none(path: str):
    if not os.path.exists(path):
        return None
    return SimpleVectorStore.from_persist_path(path)

try:
    store = SimpleVectorStore.from_persist_path(p)
except ValueError as e:
    if "No existing" in str(e):
        store = SimpleVectorStore()  # first run: build fresh
    else:
        raise

When it happens

Trigger: Calling `SimpleVectorStore.from_persist_path("./storage/vector_store.json")` (or a custom path/fs) when the file was never created — e.g. running a query-only script before the ingestion script, wrong working directory, typo in the path, or a remote fsspec filesystem where the file lives elsewhere.

Common situations: Splitting ingestion and querying into separate processes where the query job runs first; relative paths resolved against a different CWD (note the default is `os.path.join(DEFAULT_PERSIST_DIR, ...)`); CI or fresh containers that mount storage incorrectly; passing an fsspec URL/path mismatch for S3/GCS filesystems.

Related errors

• Cannot filter stores that were persisted without metadata. P
• Node content not found in metadata dict.
• SimpleVectorStore does not store nodes directly.
• Invalid query mode: {query.mode}
• Vector Store only supports exact match filters. Please use E

AI-assisted analysis of run-llama/llama_index@afd0fef371 (2026-08-15). Data as JSON: /api/errors/fc6572e3e2d53027. Report an issue: GitHub.