Skip to content

feat(py): MiddlewareV2 Implementation #4903

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

Draft
huangjeff5 wants to merge 13 commits into
base: jh-redesign
Choose a base branch
from
+2,103 −1,031
Draft

feat(py): MiddlewareV2 Implementation #4903

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
13 commits
Select commit Hold shift + click to select a range
8d08959
feat(py): updated middleware features
huangjeff5 Mar 11, 2026
840897e
fix issues
huangjeff5 Mar 11, 2026
ee06ccc
reorganize
huangjeff5 Mar 11, 2026
53b6e97
format
huangjeff5 Mar 11, 2026
35987d0
fix ruff
huangjeff5 Mar 11, 2026
403dccd
implement middleware v2
huangjeff5 Mar 11, 2026
471e447
fix type issues, improve demo
huangjeff5 Mar 11, 2026
1af1a62
format
huangjeff5 Mar 11, 2026
20e0ff3
fix
huangjeff5 Mar 11, 2026
5d048f3
cleanup
huangjeff5 Mar 11, 2026
f3c85c4
improve naming
huangjeff5 Mar 11, 2026
e696173
format
huangjeff5 Mar 11, 2026
aee2f4e
organize middleware files more scalably
huangjeff5 Mar 11, 2026
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
2 changes: 1 addition & 1 deletion py/README.md
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -443,7 +443,7 @@ examples of any feature:
| `framework-prompt-demo` | Prompt, Dotprompt | Advanced prompt templates |
| `framework-format-demo` | Structured Output | JSON/enum output formatting |
| `framework-context-demo` | Context Provider, Flow | Auth context in flows |
| `framework-middleware-demo` | Middleware | Model-level request/response hooks |
| `framework-middleware-demo` | Middleware | Built-in retry, fallback, and custom middleware |
| `framework-evaluator-demo` | Evaluator | Custom evaluation metrics |
| `framework-restaurant-demo` | Flow, Tool, Prompt | Multi-step agent with tools |
| `framework-dynamic-tools-demo` | Tool, Dynamic Action Provider | Runtime tool registration |
Expand Down
26 changes: 13 additions & 13 deletions py/packages/genkit/src/genkit/_ai/_aio.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -50,7 +50,6 @@
Message,
ModelConfig,
ModelFn,
ModelMiddleware,
ModelResponse,
ModelResponseChunk,
define_model,
Expand Down Expand Up @@ -92,6 +91,7 @@
from genkit._core._error import GenkitError
from genkit._core._flow import define_flow
from genkit._core._logger import get_logger
from genkit._core._middleware._base import BaseMiddleware
from genkit._core._plugin import Plugin
from genkit._core._reflection import ReflectionServer, ServerSpec, create_reflection_asgi_app
from genkit._core._registry import Registry
Expand Down Expand Up @@ -423,7 +423,7 @@ def define_prompt( # pyrefly:ignore[inconsistent-overload]
metadata: dict[str, object] | None = None,
tools: list[str] | None = None,
tool_choice: ToolChoice | None = None,
use: list[ModelMiddleware] | None = None,
use: list[BaseMiddleware] | None = None,
docs: list[Document] | None = None,
*,
input_schema: type[InputT],
Expand Down Expand Up @@ -452,7 +452,7 @@ def define_prompt( # pyrefly:ignore[inconsistent-overload]
metadata: dict[str, object] | None = None,
tools: list[str] | None = None,
tool_choice: ToolChoice | None = None,
use: list[ModelMiddleware] | None = None,
use: list[BaseMiddleware] | None = None,
docs: list[Document] | None = None,
*,
input_schema: type[InputT],
Expand Down Expand Up @@ -480,7 +480,7 @@ def define_prompt( # pyrefly:ignore[inconsistent-overload]
metadata: dict[str, object] | None = None,
tools: list[str] | None = None,
tool_choice: ToolChoice | None = None,
use: list[ModelMiddleware] | None = None,
use: list[BaseMiddleware] | None = None,
docs: list[Document] | None = None,
*,
output_schema: type[OutputT],
Expand Down Expand Up @@ -509,7 +509,7 @@ def define_prompt(
metadata: dict[str, object] | None = None,
tools: list[str] | None = None,
tool_choice: ToolChoice | None = None,
use: list[ModelMiddleware] | None = None,
use: list[BaseMiddleware] | None = None,
docs: list[Document] | None = None,
) -> ExecutablePrompt[Any, Any]: ...

Expand All @@ -535,7 +535,7 @@ def define_prompt( # pyright: ignore[reportInconsistentOverload]
metadata: dict[str, object] | None = None,
tools: list[str] | None = None,
tool_choice: ToolChoice | None = None,
use: list[ModelMiddleware] | None = None,
use: list[BaseMiddleware] | None = None,
docs: list[Document] | None = None,
) -> ExecutablePrompt[Any, Any]:
"""Register a prompt template."""
Expand Down Expand Up @@ -781,7 +781,7 @@ async def generate(
output_content_type: str | None = None,
output_instructions: str | None = None,
output_constrained: bool | None = None,
use: list[ModelMiddleware] | None = None,
use: list[BaseMiddleware] | None = None,
docs: list[Document] | None = None,
) -> ModelResponse[OutputT]: ...

Expand All @@ -806,7 +806,7 @@ async def generate(
output_content_type: str | None = None,
output_instructions: str | None = None,
output_constrained: bool | None = None,
use: list[ModelMiddleware] | None = None,
use: list[BaseMiddleware] | None = None,
docs: list[Document] | None = None,
) -> ModelResponse[Any]: ...

Expand All @@ -829,7 +829,7 @@ async def generate(
output_content_type: str | None = None,
output_instructions: str | None = None,
output_constrained: bool | None = None,
use: list[ModelMiddleware] | None = None,
use: list[BaseMiddleware] | None = None,
docs: list[Document] | None = None,
) -> ModelResponse[Any]:
"""Generate text or structured data using a language model."""
Expand Down Expand Up @@ -880,7 +880,7 @@ def generate_stream(
output_content_type: str | None = None,
output_instructions: str | None = None,
output_constrained: bool | None = None,
use: list[ModelMiddleware] | None = None,
use: list[BaseMiddleware] | None = None,
docs: list[Document] | None = None,
timeout: float | None = None,
) -> ModelStreamResponse[OutputT]: ...
Expand All @@ -905,7 +905,7 @@ def generate_stream(
output_content_type: str | None = None,
output_instructions: str | None = None,
output_constrained: bool | None = None,
use: list[ModelMiddleware] | None = None,
use: list[BaseMiddleware] | None = None,
docs: list[Document] | None = None,
timeout: float | None = None,
) -> ModelStreamResponse[Any]: ...
Expand All @@ -928,7 +928,7 @@ def generate_stream(
output_content_type: str | None = None,
output_instructions: str | None = None,
output_constrained: bool | None = None,
use: list[ModelMiddleware] | None = None,
use: list[BaseMiddleware] | None = None,
docs: list[Document] | None = None,
timeout: float | None = None,
) -> ModelStreamResponse[Any]:
Expand Down Expand Up @@ -1171,7 +1171,7 @@ async def generate_operation(
output_content_type: str | None = None,
output_instructions: str | None = None,
output_constrained: bool | None = None,
use: list[ModelMiddleware] | None = None,
use: list[BaseMiddleware] | None = None,
docs: list[Document] | None = None,
) -> Operation:
"""Generate content using a long-running model, returning an Operation to poll."""
Expand Down
Loading