LLM-Bridge 2.1.0__tar.gz → 2.1.2__tar.gz

{llm_bridge-2.1.0 → llm_bridge-2.1.2}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: LLM-Bridge
-Version: 2.1.0
+Version: 2.1.2
 Summary: A library that wraps multiple LLM providers into a consistent API while using each provider's native SDK internally, supporting multimodal I/O, file processing, and stream output.
 Author-email: windsnow1025 <windsnow1025@gmail.com>
 License-Expression: MIT

{llm_bridge-2.1.0 → llm_bridge-2.1.2}/llm_bridge/client/chat_client.py

@@ -1,4 +1,4 @@
-from typing import AsyncGenerator
+from collections.abc import AsyncGenerator
 
 from llm_bridge.type.chat_response import ChatResponse
 

{llm_bridge-2.1.0 → llm_bridge-2.1.2}/llm_bridge/client/implementations/claude/stream_claude_client.py

@@ -1,6 +1,6 @@
 import logging
 import re
-from typing import AsyncGenerator
+from collections.abc import AsyncGenerator
 
 import httpx
 from fastapi import HTTPException

{llm_bridge-2.1.0 → llm_bridge-2.1.2}/llm_bridge/client/implementations/gemini/gemini_response_handler.py

@@ -1,5 +1,4 @@
 import mimetypes
-from typing import Optional
 
 from google.genai import types
 from google.genai.types import Part
@@ -22,7 +21,7 @@ class GeminiResponseHandler:
         code: str = ""
         code_output: str = ""
         files: list[File] = []
-        display: Optional[str] = None
+        display: str | None = None
         input_tokens, cumulative_output_tokens = await count_gemini_tokens(response)
 
         parts: list[Part] = []

{llm_bridge-2.1.0 → llm_bridge-2.1.2}/llm_bridge/client/implementations/gemini/stream_gemini_client.py

@@ -1,6 +1,6 @@
 import logging
 import re
-from typing import AsyncGenerator, AsyncIterator
+from collections.abc import AsyncGenerator, AsyncIterator
 
 import httpx
 from fastapi import HTTPException

{llm_bridge-2.1.0 → llm_bridge-2.1.2}/llm_bridge/client/implementations/openai_completion/stream_openai_completion_client.py

@@ -1,6 +1,6 @@
 import logging
 import re
-from typing import AsyncGenerator
+from collections.abc import AsyncGenerator
 
 import httpx
 import openai

{llm_bridge-2.1.0 → llm_bridge-2.1.2}/llm_bridge/client/implementations/openai_responses/stream_openai_responses_client.py

@@ -1,6 +1,6 @@
 import logging
 import re
-from typing import AsyncGenerator
+from collections.abc import AsyncGenerator
 
 import httpx
 import openai

{llm_bridge-2.1.0 → llm_bridge-2.1.2}/llm_bridge/client/implementations/xai/stream_xai_client.py

@@ -1,6 +1,6 @@
 import logging
 import re
-from typing import AsyncGenerator
+from collections.abc import AsyncGenerator
 
 from fastapi import HTTPException
 from xai_sdk.aio.chat import Chat

{llm_bridge-2.1.0 → llm_bridge-2.1.2}/llm_bridge/client/model_client/claude_client.py

@@ -1,4 +1,4 @@
-from typing import AsyncGenerator
+from collections.abc import AsyncGenerator
 
 import anthropic
 from anthropic import Omit
@@ -17,9 +17,9 @@ class ClaudeClient(ChatClient):
             model: str,
             messages: list[ClaudeMessage],
             system: str,
-            client: anthropic.AsyncAnthropic,
             max_tokens: int,
-            betas: list[AnthropicBetaParam],
+            client: anthropic.AsyncAnthropic,
+            betas: list[AnthropicBetaParam] | Omit,
             tools: list[BetaToolUnionParam],
             cache_control: BetaCacheControlEphemeralParam,
             thinking: BetaThinkingConfigParam | Omit,
@@ -28,8 +28,8 @@ class ClaudeClient(ChatClient):
         self.model = model
         self.messages = messages
         self.system = system
-        self.client = client
         self.max_tokens = max_tokens
+        self.client = client
         self.betas = betas
         self.tools = tools
         self.cache_control = cache_control

{llm_bridge-2.1.0 → llm_bridge-2.1.2}/llm_bridge/client/model_client/gemini_client.py

@@ -1,4 +1,4 @@
-from typing import AsyncGenerator
+from collections.abc import AsyncGenerator
 
 from google import genai
 from google.genai import types

{llm_bridge-2.1.0 → llm_bridge-2.1.2}/llm_bridge/client/model_client/openai_completion_client.py

@@ -1,4 +1,5 @@
-from typing import AsyncGenerator, Literal
+from collections.abc import AsyncGenerator
+from typing import Literal
 
 import openai.lib.azure
 import openai

{llm_bridge-2.1.0 → llm_bridge-2.1.2}/llm_bridge/client/model_client/openai_responses_client.py

@@ -1,4 +1,4 @@
-from typing import AsyncGenerator, Iterable
+from collections.abc import AsyncGenerator, Iterable
 
 import openai.lib.azure
 import openai

{llm_bridge-2.1.0 → llm_bridge-2.1.2}/llm_bridge/client/model_client/xai_client.py

@@ -1,4 +1,4 @@
-from typing import AsyncGenerator
+from collections.abc import AsyncGenerator
 
 import xai_sdk
 from xai_sdk.proto import chat_pb2

{llm_bridge-2.1.0 → llm_bridge-2.1.2}/llm_bridge/logic/chat_generate/model_client_factory/claude_client_factory.py

@@ -59,13 +59,7 @@ async def create_claude_client(
     if thought:
         thinking = BetaThinkingConfigAdaptiveParam(type="adaptive", display="summarized")
 
-    betas: list[AnthropicBetaParam] = [
-        "context-1m-2025-08-07",
-        "output-128k-2025-02-19",
-        "code-execution-2025-08-25",
-        "files-api-2025-04-14",
-        "structured-outputs-2025-11-13",
-    ]
+    betas: list[AnthropicBetaParam] | Omit = omit
 
     tools: list[BetaToolUnionParam] = []
     if web_search:
@@ -101,8 +95,8 @@ async def create_claude_client(
             model=model,
             messages=claude_messages,
             system=system,
-            client=client,
             max_tokens=max_tokens,
+            client=client,
             betas=betas,
             tools=tools,
             cache_control=cache_control,
@@ -114,8 +108,8 @@ async def create_claude_client(
             model=model,
             messages=claude_messages,
             system=system,
-            client=client,
             max_tokens=max_tokens,
+            client=client,
             betas=betas,
             tools=tools,
             cache_control=cache_control,

llm_bridge-2.1.2/llm_bridge/type/chat_response.py

@@ -0,0 +1,22 @@
+from dataclasses import dataclass
+
+
+@dataclass
+class File:
+    name: str
+    data: str
+    type: str
+
+
+@dataclass
+class ChatResponse:
+    text: str | None = None
+    audio: str | None = None
+    thought: str | None = None
+    code: str | None = None
+    code_output: str | None = None
+    files: list[File] | None = None
+    display: str | None = None
+    error: str | None = None
+    input_tokens: int | None = 0
+    output_tokens: int | None = 0

{llm_bridge-2.1.0 → llm_bridge-2.1.2}/llm_bridge/type/message.py

@@ -1,5 +1,5 @@
 from dataclasses import dataclass
-from enum import Enum, StrEnum
+from enum import StrEnum
 
 
 class Role(StrEnum):

{llm_bridge-2.1.0 → llm_bridge-2.1.2}/llm_bridge/type/model_message/claude_message.py

@@ -1,5 +1,4 @@
 from enum import StrEnum
-from typing import TypeAlias
 
 from anthropic.types import MessageParam, TextBlockParam, ImageBlockParam, DocumentBlockParam
 
@@ -9,9 +8,9 @@ class ClaudeRole(StrEnum):
     Assistant = "assistant"
 
 
-ClaudeContent: TypeAlias = TextBlockParam | ImageBlockParam | DocumentBlockParam
+type ClaudeContent = TextBlockParam | ImageBlockParam | DocumentBlockParam
 
-ClaudeMessage: TypeAlias = MessageParam
+type ClaudeMessage = MessageParam
 
 # class MessageParam(TypedDict, total=False):
 #     content: Required[

{llm_bridge-2.1.0 → llm_bridge-2.1.2}/llm_bridge/type/model_message/gemini_message.py

@@ -1,5 +1,4 @@
 from enum import StrEnum
-from typing import TypeAlias
 
 from google.genai import types
 
@@ -9,6 +8,6 @@ class GeminiRole(StrEnum):
     Model = "model"
 
 
-GeminiContent: TypeAlias = types.Part
+type GeminiContent = types.Part
 
-GeminiMessage: TypeAlias = types.Content
+type GeminiMessage = types.Content

llm_bridge-2.1.2/llm_bridge/type/model_message/openai_completion_message.py

@@ -0,0 +1,12 @@
+from openai.types.chat import ChatCompletionSystemMessageParam, ChatCompletionUserMessageParam, \
+    ChatCompletionAssistantMessageParam
+from openai.types.chat.chat_completion_content_part_param import ChatCompletionContentPartParam
+
+from llm_bridge.type.message import Role
+
+type OpenAICompletionRole = Role
+
+type OpenAICompletionContent = ChatCompletionContentPartParam
+
+# type OpenAICompletionMessage = ChatCompletionMessageParam
+type OpenAICompletionMessage = ChatCompletionSystemMessageParam | ChatCompletionUserMessageParam | ChatCompletionAssistantMessageParam

llm_bridge-2.1.2/llm_bridge/type/model_message/openai_responses_message.py

@@ -0,0 +1,10 @@
+from openai.types.responses import EasyInputMessageParam, ResponseOutputMessageParam, ResponseInputContentParam, \
+    ResponseOutputTextParam
+
+from llm_bridge.type.message import Role
+
+type OpenAIResponsesRole = Role
+
+type OpenAIResponsesContent = ResponseInputContentParam | ResponseOutputTextParam
+
+type OpenAIResponsesMessage = EasyInputMessageParam | ResponseOutputMessageParam

llm_bridge-2.1.2/llm_bridge/type/model_message/xai_message.py

@@ -0,0 +1,5 @@
+from xai_sdk.chat import chat_pb2
+
+type XAIContent = chat_pb2.Content
+
+type XAIMessage = chat_pb2.Message

{llm_bridge-2.1.0 → llm_bridge-2.1.2}/pyproject.toml

@@ -4,7 +4,7 @@ build-backend = "hatchling.build"
 
 [project]
 name = "LLM-Bridge"
-version = "2.1.0"
+version = "2.1.2"
 dependencies = [
     "fastapi",
     "httpx",

{llm_bridge-2.1.0 → llm_bridge-2.1.2}/usage/batch/batch.py

@@ -55,13 +55,11 @@ async def test_model(
 
     for attempt in range(1, MaxRetries + 1):
         try:
-            latency = await asyncio.wait_for(
-                measure_first_chunk_latency(api_type, model, config),
-                timeout=TimeoutSeconds,
-            )
+            async with asyncio.timeout(TimeoutSeconds):
+                latency = await measure_first_chunk_latency(api_type, model, config)
             print(f"  [{attempt}/{MaxRetries}] OK      {api_type:28s} {model:35s} [{config.name:10s}] {latency:8.3f}s")
             break
-        except asyncio.TimeoutError:
+        except TimeoutError:
             error = f"TIMEOUT ({TimeoutSeconds}s)"
             print(f"  [{attempt}/{MaxRetries}] TIMEOUT {api_type:28s} {model:35s} [{config.name:10s}]")
         except Exception as e:

{llm_bridge-2.1.0 → llm_bridge-2.1.2}/usage/workflow.py

@@ -1,4 +1,5 @@
-from typing import AsyncGenerator, Any
+from collections.abc import AsyncGenerator
+from typing import Any
 
 from llm_bridge import *
 

{llm_bridge-2.1.0 → llm_bridge-2.1.2}/uv.lock

@@ -748,7 +748,7 @@ wheels = [
 
 [[package]]
 name = "llm-bridge"
-version = "2.1.0"
+version = "2.1.2"
 source = { editable = "." }
 dependencies = [
     { name = "anthropic" },

llm_bridge-2.1.0/MANIFEST.in

@@ -1 +0,0 @@
-include llm_bridge/resources/*.json

llm_bridge-2.1.0/llm_bridge/type/chat_response.py

@@ -1,23 +0,0 @@
-from dataclasses import dataclass
-from typing import Optional
-
-
-@dataclass
-class File:
-    name: str
-    data: str
-    type: str
-
-
-@dataclass
-class ChatResponse:
-    text: Optional[str] = None
-    audio: Optional[str] = None
-    thought: Optional[str] = None
-    code: Optional[str] = None
-    code_output: Optional[str] = None
-    files: Optional[list[File]] = None
-    display: Optional[str] = None
-    error: Optional[str] = None
-    input_tokens: Optional[int] = 0
-    output_tokens: Optional[int] = 0

llm_bridge-2.1.0/llm_bridge/type/model_message/openai_completion_message.py

@@ -1,14 +0,0 @@
-from typing import TypeAlias
-
-from openai.types.chat import ChatCompletionSystemMessageParam, ChatCompletionUserMessageParam, \
-    ChatCompletionAssistantMessageParam
-from openai.types.chat.chat_completion_content_part_param import ChatCompletionContentPartParam
-
-from llm_bridge.type.message import Role
-
-OpenAICompletionRole: TypeAlias = Role
-
-OpenAICompletionContent: TypeAlias = ChatCompletionContentPartParam
-
-# OpenAICompletionMessage: TypeAlias = ChatCompletionMessageParam
-OpenAICompletionMessage: TypeAlias = ChatCompletionSystemMessageParam | ChatCompletionUserMessageParam | ChatCompletionAssistantMessageParam

llm_bridge-2.1.0/llm_bridge/type/model_message/openai_responses_message.py

@@ -1,12 +0,0 @@
-from typing import TypeAlias
-
-from openai.types.responses import EasyInputMessageParam, ResponseOutputMessageParam, ResponseInputContentParam, \
-    ResponseOutputTextParam
-
-from llm_bridge.type.message import Role
-
-OpenAIResponsesRole: TypeAlias = Role
-
-OpenAIResponsesContent: TypeAlias = ResponseInputContentParam | ResponseOutputTextParam
-
-OpenAIResponsesMessage: TypeAlias = EasyInputMessageParam | ResponseOutputMessageParam

llm_bridge-2.1.0/llm_bridge/type/model_message/xai_message.py

@@ -1,7 +0,0 @@
-from typing import TypeAlias
-
-from xai_sdk.chat import chat_pb2
-
-XAIContent: TypeAlias = chat_pb2.Content
-
-XAIMessage: TypeAlias = chat_pb2.Message