Skip to content

Add OpenAIBatchChatGenerator for processing batch API requests (#8482) #12268

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

Open
DhanushPillay wants to merge 5 commits into
base: main
Choose a base branch
from
+1,214 −89
Open

Add OpenAIBatchChatGenerator for processing batch API requests (#8482) #12268

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
5 commits
Select commit Hold shift + click to select a range
88bdf07
Add OpenAIBatchChatGenerator for processing batch API requests
DhanushPillay Aug 7, 2026
399d968
fix: remove unused type: ignore comments and fix test kwargs for mypy
DhanushPillay Aug 27, 2026
ddce911
style: apply ruff formatting
DhanushPillay Aug 27, 2026
37b8c12
fix: update test assertions to match current OpenAI SDK token detail …
DhanushPillay Aug 27, 2026
db6eac2
style: apply ruff formatting
DhanushPillay Aug 27, 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
10 changes: 2 additions & 8 deletions haystack/components/embedders/azure_document_embedder.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -173,10 +173,7 @@ def warm_up(self) -> None:
# openai>=3 annotates http_client as httpx2, but legacy httpx clients are supported at runtime.
# https://github.com/openai/openai-python/blob/main/httpx2.md
http_client = init_http_client(self.http_client_kwargs, async_client=False)
self.client = AzureOpenAI(
http_client=http_client, # type: ignore[arg-type]
**self._client_kwargs(),
)
self.client = AzureOpenAI(http_client=http_client, **self._client_kwargs())

async def warm_up_async(self) -> None: # noqa: RUF029
"""
Expand All @@ -186,10 +183,7 @@ async def warm_up_async(self) -> None: # noqa: RUF029
# openai>=3 annotates http_client as httpx2, but legacy httpx clients are supported at runtime.
# https://github.com/openai/openai-python/blob/main/httpx2.md
http_client = init_http_client(self.http_client_kwargs, async_client=True)
self.async_client = AsyncAzureOpenAI(
http_client=http_client, # type: ignore[arg-type]
**self._client_kwargs(),
)
self.async_client = AsyncAzureOpenAI(http_client=http_client, **self._client_kwargs())

def close(self) -> None:
"""
Expand Down
10 changes: 2 additions & 8 deletions haystack/components/embedders/azure_text_embedder.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -154,10 +154,7 @@ def warm_up(self) -> None:
# openai>=3 annotates http_client as httpx2, but legacy httpx clients are supported at runtime.
# https://github.com/openai/openai-python/blob/main/httpx2.md
http_client = init_http_client(self.http_client_kwargs, async_client=False)
self.client = AzureOpenAI(
http_client=http_client, # type: ignore[arg-type]
**self._client_kwargs(),
)
self.client = AzureOpenAI(http_client=http_client, **self._client_kwargs())

async def warm_up_async(self) -> None: # noqa: RUF029
"""
Expand All @@ -167,10 +164,7 @@ async def warm_up_async(self) -> None: # noqa: RUF029
# openai>=3 annotates http_client as httpx2, but legacy httpx clients are supported at runtime.
# https://github.com/openai/openai-python/blob/main/httpx2.md
http_client = init_http_client(self.http_client_kwargs, async_client=True)
self.async_client = AsyncAzureOpenAI(
http_client=http_client, # type: ignore[arg-type]
**self._client_kwargs(),
)
self.async_client = AsyncAzureOpenAI(http_client=http_client, **self._client_kwargs())

def close(self) -> None:
"""
Expand Down
10 changes: 2 additions & 8 deletions haystack/components/embedders/openai_document_embedder.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -146,10 +146,7 @@ def warm_up(self) -> None:
# openai>=3 annotates http_client as httpx2, but legacy httpx clients are supported at runtime.
# https://github.com/openai/openai-python/blob/main/httpx2.md
http_client = init_http_client(self.http_client_kwargs, async_client=False)
self.client = OpenAI(
http_client=http_client, # type: ignore[arg-type]
**self._client_kwargs(),
)
self.client = OpenAI(http_client=http_client, **self._client_kwargs())

async def warm_up_async(self) -> None: # noqa: RUF029
"""
Expand All @@ -159,10 +156,7 @@ async def warm_up_async(self) -> None: # noqa: RUF029
# openai>=3 annotates http_client as httpx2, but legacy httpx clients are supported at runtime.
# https://github.com/openai/openai-python/blob/main/httpx2.md
http_client = init_http_client(self.http_client_kwargs, async_client=True)
self.async_client = AsyncOpenAI(
http_client=http_client, # type: ignore[arg-type]
**self._client_kwargs(),
)
self.async_client = AsyncOpenAI(http_client=http_client, **self._client_kwargs())

def close(self) -> None:
"""
Expand Down
10 changes: 2 additions & 8 deletions haystack/components/embedders/openai_text_embedder.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -121,10 +121,7 @@ def warm_up(self) -> None:
# openai>=3 annotates http_client as httpx2, but legacy httpx clients are supported at runtime.
# https://github.com/openai/openai-python/blob/main/httpx2.md
http_client = init_http_client(self.http_client_kwargs, async_client=False)
self.client = OpenAI(
http_client=http_client, # type: ignore[arg-type]
**self._client_kwargs(),
)
self.client = OpenAI(http_client=http_client, **self._client_kwargs())

async def warm_up_async(self) -> None: # noqa: RUF029
"""
Expand All @@ -134,10 +131,7 @@ async def warm_up_async(self) -> None: # noqa: RUF029
# openai>=3 annotates http_client as httpx2, but legacy httpx clients are supported at runtime.
# https://github.com/openai/openai-python/blob/main/httpx2.md
http_client = init_http_client(self.http_client_kwargs, async_client=True)
self.async_client = AsyncOpenAI(
http_client=http_client, # type: ignore[arg-type]
**self._client_kwargs(),
)
self.async_client = AsyncOpenAI(http_client=http_client, **self._client_kwargs())

def close(self) -> None:
"""
Expand Down
2 changes: 2 additions & 0 deletions haystack/components/generators/chat/__init__.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -9,6 +9,7 @@

_import_structure = {
"openai": ["OpenAIChatGenerator"],
"openai_batch": ["OpenAIBatchChatGenerator"],
"openai_responses": ["OpenAIResponsesChatGenerator"],
"azure": ["AzureOpenAIChatGenerator"],
"azure_responses": ["AzureOpenAIResponsesChatGenerator"],
Expand All @@ -24,6 +25,7 @@
from .llm import LLM as LLM
from .mock import MockChatGenerator as MockChatGenerator
from .openai import OpenAIChatGenerator as OpenAIChatGenerator
from .openai_batch import OpenAIBatchChatGenerator as OpenAIBatchChatGenerator
from .openai_responses import OpenAIResponsesChatGenerator as OpenAIResponsesChatGenerator

else:
Expand Down
10 changes: 2 additions & 8 deletions haystack/components/generators/chat/azure.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -274,10 +274,7 @@ def warm_up(self) -> None:
# openai>=3 annotates http_client as httpx2, but legacy httpx clients are supported at runtime.
# https://github.com/openai/openai-python/blob/main/httpx2.md
http_client = init_http_client(self.http_client_kwargs, async_client=False)
self.client = AzureOpenAI(
http_client=http_client, # type: ignore[arg-type]
**self._client_kwargs(),
)
self.client = AzureOpenAI(http_client=http_client, **self._client_kwargs())

async def warm_up_async(self) -> None: # noqa: RUF029
"""
Expand All @@ -288,10 +285,7 @@ async def warm_up_async(self) -> None: # noqa: RUF029
# openai>=3 annotates http_client as httpx2, but legacy httpx clients are supported at runtime.
# https://github.com/openai/openai-python/blob/main/httpx2.md
http_client = init_http_client(self.http_client_kwargs, async_client=True)
self.async_client = AsyncAzureOpenAI(
http_client=http_client, # type: ignore[arg-type]
**self._client_kwargs(),
)
self.async_client = AsyncAzureOpenAI(http_client=http_client, **self._client_kwargs())

def close(self) -> None:
"""
Expand Down
10 changes: 2 additions & 8 deletions haystack/components/generators/chat/openai.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -240,10 +240,7 @@ def warm_up(self) -> None:
# openai>=3 annotates http_client as httpx2, but legacy httpx clients are supported at runtime.
# https://github.com/openai/openai-python/blob/main/httpx2.md
http_client = init_http_client(self.http_client_kwargs, async_client=False)
self.client = OpenAI(
http_client=http_client, # type: ignore[arg-type]
**self._client_kwargs(),
)
self.client = OpenAI(http_client=http_client, **self._client_kwargs())

async def warm_up_async(self) -> None: # noqa: RUF029
"""
Expand All @@ -254,10 +251,7 @@ async def warm_up_async(self) -> None: # noqa: RUF029
# openai>=3 annotates http_client as httpx2, but legacy httpx clients are supported at runtime.
# https://github.com/openai/openai-python/blob/main/httpx2.md
http_client = init_http_client(self.http_client_kwargs, async_client=True)
self.async_client = AsyncOpenAI(
http_client=http_client, # type: ignore[arg-type]
**self._client_kwargs(),
)
self.async_client = AsyncOpenAI(http_client=http_client, **self._client_kwargs())

def close(self) -> None:
"""
Expand Down
Loading
Loading