run-llama/llama_index · warning · DeprecationWarning

PandasQueryEngine has been moved to `llama-index-experimenta

Error message

PandasQueryEngine has been moved to `llama-index-experimental`.
`pip install llama-index-experimental`
`from llama_index.experimental.query_engine import PandasQueryEngine`
Note that the PandasQueryEngine allows for arbitrary code execution, 
and should be used in a secure environment.

What it means

PandasIndex (and GPTPandasIndex) were moved out of llama-index-core into llama-index-experimental because PandasQueryEngine generates and executes arbitrary pandas code. The stub class raises DeprecationWarning in __init__ so old imports keep working but instantiation fails with instructions for the new package.

Source

Thrown at llama-index-core/llama_index/core/indices/struct_store/pandas.py:16

"""
Pandas csv structured store.

DEPRECATED: Please use :class:`PandasQueryEngine` in `llama-index-experimental` instead.
"""

from typing import Any


class PandasIndex:
    def __init__(
        self,
        *args: Any,
        **kwargs: Any,
    ) -> None:
        raise DeprecationWarning(
            "PandasQueryEngine has been moved to `llama-index-experimental`.\n"
            "`pip install llama-index-experimental`\n"
            "`from llama_index.experimental.query_engine import PandasQueryEngine`\n"
            "Note that the PandasQueryEngine allows for arbitrary code execution, \n"
            "and should be used in a secure environment."
        )


# Legacy
GPTPandasIndex = PandasIndex

View on GitHub (pinned to afd0fef371)

Solutions

pip install llama-index-experimental and change the import to from llama_index.experimental.query_engine import PandasQueryEngine
Use PandasQueryEngine(df, ...) instead of building a PandasIndex directly
Run it only in a sandboxed/secure environment — it executes LLM-generated code by design

Example fix

# before
from llama_index.core.indices.struct_store.pandas import PandasIndex, GPTPandasIndex
index = PandasIndex(df=df)  # raises DeprecationWarning

# after
pip install llama-index-experimental
from llama_index.experimental.query_engine import PandasQueryEngine
query_engine = PandasQueryEngine(df=df)

Defensive patterns

Strategy: fallback

Validate before calling

try:
    from llama_index.experimental.query_engine import PandasQueryEngine
    HAS_EXPERIMENTAL = True
except ImportError:
    HAS_EXPERIMENTAL = False

if not HAS_EXPERIMENTAL:
    raise RuntimeError('pip install llama-index-experimental for PandasQueryEngine')

Type guard

def pandas_engine_available() -> bool:
    try:
        from llama_index.experimental.query_engine import PandasQueryEngine  # noqa: F401
        return True
    except ImportError:
        return False

Try / catch

try:
    index = PandasIndex(df=df)
except DeprecationWarning as e:
    if 'llama-index-experimental' in str(e):
        from llama_index.experimental.query_engine import PandasQueryEngine
        query_engine = PandasQueryEngine(df=df)
    else:
        raise

Prevention

After any llama-index upgrade, grep for imports under core that moved to integration/experimental packages
Run PandasQueryEngine only in sandboxed environments since it executes generated code

When it happens

Trigger: Calling PandasIndex(...) or GPTPandasIndex(...) on llama-index-core after the removal (roughly v0.10.44+); running old notebooks/scripts built before the move.

Common situations: Upgrading llama-index past the deprecation boundary; tutorials showing the old core import path; CI suddenly failing after a dependency bump.

Related errors

AI-assisted analysis of run-llama/llama_index@afd0fef371 (2026-08-15). Data as JSON: /api/errors/48e57b1732c91188. Report an issue: GitHub.