Skip to content

feat: support string literal inputs for AI functions #2152

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Merged
merged 4 commits into from
Oct 8, 2025
Merged

feat: support string literal inputs for AI functions #2152

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
4 commits
Select commit Hold shift + click to select a range
00dd022
feat: support string literal inputs for AI functions
sycai Oct 8, 2025
6d2f5a2
polish code
sycai Oct 8, 2025
59c3339
update pydoc
sycai Oct 8, 2025
e5108bf
Merge branch 'main' into sycai_ai_string_input
sycai Oct 8, 2025
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
20 changes: 12 additions & 8 deletions bigframes/bigquery/_operations/ai.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -28,6 +28,7 @@
from bigframes.operations import ai_ops, output_schemas

PROMPT_TYPE = Union[
str,
series.Series,
pd.Series,
List[Union[str, series.Series, pd.Series]],
Expand Down Expand Up @@ -73,7 +74,7 @@ def generate(
dtype: struct<is_herbivore: bool, number_of_legs: int64, full_response: extension<dbjson<JSONArrowType>>, status: string>[pyarrow]

Args:
prompt (Series | List[str|Series] | Tuple[str|Series, ...]):
prompt (str | Series | List[str|Series] | Tuple[str|Series, ...]):
A mixture of Series and string literals that specifies the prompt to send to the model. The Series can be BigFrames Series
or pandas Series.
connection_id (str, optional):
Expand Down Expand Up @@ -165,7 +166,7 @@ def generate_bool(
Name: result, dtype: boolean

Args:
prompt (Series | List[str|Series] | Tuple[str|Series, ...]):
prompt (str | Series | List[str|Series] | Tuple[str|Series, ...]):
A mixture of Series and string literals that specifies the prompt to send to the model. The Series can be BigFrames Series
or pandas Series.
connection_id (str, optional):
Expand Down Expand Up @@ -240,7 +241,7 @@ def generate_int(
Name: result, dtype: Int64

Args:
prompt (Series | List[str|Series] | Tuple[str|Series, ...]):
prompt (str | Series | List[str|Series] | Tuple[str|Series, ...]):
A mixture of Series and string literals that specifies the prompt to send to the model. The Series can be BigFrames Series
or pandas Series.
connection_id (str, optional):
Expand Down Expand Up @@ -315,7 +316,7 @@ def generate_double(
Name: result, dtype: Float64

Args:
prompt (Series | List[str|Series] | Tuple[str|Series, ...]):
prompt (str | Series | List[str|Series] | Tuple[str|Series, ...]):
A mixture of Series and string literals that specifies the prompt to send to the model. The Series can be BigFrames Series
or pandas Series.
connection_id (str, optional):
Expand Down Expand Up @@ -386,7 +387,7 @@ def if_(
dtype: string

Args:
prompt (Series | List[str|Series] | Tuple[str|Series, ...]):
prompt (str | Series | List[str|Series] | Tuple[str|Series, ...]):
A mixture of Series and string literals that specifies the prompt to send to the model. The Series can be BigFrames Series
or pandas Series.
connection_id (str, optional):
Expand Down Expand Up @@ -433,7 +434,7 @@ def classify(
[2 rows x 2 columns]

Args:
input (Series | List[str|Series] | Tuple[str|Series, ...]):
input (str | Series | List[str|Series] | Tuple[str|Series, ...]):
A mixture of Series and string literals that specifies the input to send to the model. The Series can be BigFrames Series
or pandas Series.
categories (tuple[str, ...] | list[str]):
Expand Down Expand Up @@ -482,7 +483,7 @@ def score(
dtype: Float64

Args:
prompt (Series | List[str|Series] | Tuple[str|Series, ...]):
prompt (str | Series | List[str|Series] | Tuple[str|Series, ...]):
A mixture of Series and string literals that specifies the prompt to send to the model. The Series can be BigFrames Series
or pandas Series.
connection_id (str, optional):
Expand Down Expand Up @@ -514,9 +515,12 @@ def _separate_context_and_series(
Input: ("str1", series1, "str2", "str3", series2)
Output: ["str1", None, "str2", "str3", None], [series1, series2]
"""
if not isinstance(prompt, (list, tuple, series.Series)):
if not isinstance(prompt, (str, list, tuple, series.Series)):
raise ValueError(f"Unsupported prompt type: {type(prompt)}")

if isinstance(prompt, str):
return [None], [series.Series([prompt])]

if isinstance(prompt, series.Series):
if prompt.dtype == dtypes.OBJ_REF_DTYPE:
# Multi-model support
Expand Down
23 changes: 23 additions & 0 deletions tests/system/small/bigquery/test_ai.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -12,6 +12,8 @@
# See the License for the specific language governing permissions and
# limitations under the License.

from unittest import mock

from packaging import version
import pandas as pd
import pyarrow as pa
Expand Down Expand Up @@ -42,6 +44,27 @@ def test_ai_function_pandas_input(session):
)


def test_ai_function_string_input(session):
with mock.patch(
"bigframes.core.global_session.get_global_session"
) as mock_get_session:
mock_get_session.return_value = session
prompt = "Is apple a fruit?"

result = bbq.ai.generate_bool(prompt, endpoint="gemini-2.5-flash")

assert _contains_no_nulls(result)
assert result.dtype == pd.ArrowDtype(
pa.struct(
(
pa.field("result", pa.bool_()),
pa.field("full_response", dtypes.JSON_ARROW_TYPE),
pa.field("status", pa.string()),
)
)
)


def test_ai_function_compile_model_params(session):
if version.Version(sqlglot.__version__) < version.Version("25.18.0"):
pytest.skip(
Expand Down