LLM-Bridge 2.0.4__tar.gz → 2.0.6__tar.gz

{llm_bridge-2.0.4 → llm_bridge-2.0.6}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: LLM-Bridge
-Version: 2.0.4
+Version: 2.0.6
 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
@@ -40,13 +40,13 @@ PyPI: [https://pypi.org/project/LLM-Bridge/](https://pypi.org/project/LLM-Bridge
 
 The features listed represent the maximum capabilities of each API type supported by LLM Bridge.
 
-| API Type              | Input Format                                     | Capabilities                                            | Output Format     |
-|-----------------------|--------------------------------------------------|---------------------------------------------------------|-------------------|
-| OpenAI Completion API | Text, Image, PDF                                 | Thinking, Structured Output                             | Text              |
-| OpenAI Responses API  | Text, Image, PDF                                 | Thinking, Web Search, Code Execution, Structured Output | Text, Image       |
-| Google GenAI          | Text, Image, PDF, Audio, Video                   | Thinking, Web Search, Code Execution, Structured Output | Text, Image, File |
-| Anthropic             | Text, Image, PDF                                 | Thinking, Web Search, Code Execution, Structured Output | Text, File        |
-| xAI                   | Text, Image, PDF, Audio, Video, docx, xlsx, pptx | Thinking, Web Search, Code Execution, Structured Output | Text              |
+| API Type              | Input Format                                     | Capabilities                                                       | Output Format     |
+|-----------------------|--------------------------------------------------|--------------------------------------------------------------------|-------------------|
+| OpenAI Completion API | Text, Image, PDF                                 | Thinking, Structured Output                                        | Text              |
+| OpenAI Responses API  | Text, Image, PDF                                 | Thinking, Web Search, Code Execution, Structured Output            | Text, Image       |
+| Google GenAI          | Text, Image, PDF, Audio, Video                   | Thinking, Web Search, Web Fetch, Code Execution, Structured Output | Text, Image, File |
+| Anthropic             | Text, Image, PDF                                 | Thinking, Web Search, Web Fetch, Code Execution, Structured Output | Text, File        |
+| xAI                   | Text, Image, PDF, Audio, Video, docx, xlsx, pptx | Thinking, Web Search, Code Execution, Structured Output            | Text              |
 
 ## Development
 

{llm_bridge-2.0.4 → llm_bridge-2.0.6}/README.md

@@ -17,13 +17,13 @@ PyPI: [https://pypi.org/project/LLM-Bridge/](https://pypi.org/project/LLM-Bridge
 
 The features listed represent the maximum capabilities of each API type supported by LLM Bridge.
 
-| API Type              | Input Format                                     | Capabilities                                            | Output Format     |
-|-----------------------|--------------------------------------------------|---------------------------------------------------------|-------------------|
-| OpenAI Completion API | Text, Image, PDF                                 | Thinking, Structured Output                             | Text              |
-| OpenAI Responses API  | Text, Image, PDF                                 | Thinking, Web Search, Code Execution, Structured Output | Text, Image       |
-| Google GenAI          | Text, Image, PDF, Audio, Video                   | Thinking, Web Search, Code Execution, Structured Output | Text, Image, File |
-| Anthropic             | Text, Image, PDF                                 | Thinking, Web Search, Code Execution, Structured Output | Text, File        |
-| xAI                   | Text, Image, PDF, Audio, Video, docx, xlsx, pptx | Thinking, Web Search, Code Execution, Structured Output | Text              |
+| API Type              | Input Format                                     | Capabilities                                                       | Output Format     |
+|-----------------------|--------------------------------------------------|--------------------------------------------------------------------|-------------------|
+| OpenAI Completion API | Text, Image, PDF                                 | Thinking, Structured Output                                        | Text              |
+| OpenAI Responses API  | Text, Image, PDF                                 | Thinking, Web Search, Code Execution, Structured Output            | Text, Image       |
+| Google GenAI          | Text, Image, PDF, Audio, Video                   | Thinking, Web Search, Web Fetch, Code Execution, Structured Output | Text, Image, File |
+| Anthropic             | Text, Image, PDF                                 | Thinking, Web Search, Web Fetch, Code Execution, Structured Output | Text, File        |
+| xAI                   | Text, Image, PDF, Audio, Video, docx, xlsx, pptx | Thinking, Web Search, Code Execution, Structured Output            | Text              |
 
 ## Development
 

{llm_bridge-2.0.4 → llm_bridge-2.0.6}/llm_bridge/client/implementations/claude/non_stream_claude_client.py

@@ -23,6 +23,7 @@ class NonStreamClaudeClient(ClaudeClient):
                 messages=serialize(self.messages),
                 betas=self.betas,
                 tools=self.tools,
+                cache_control=self.cache_control,
                 thinking=self.thinking,
                 output_config=self.output_config,
             )

{llm_bridge-2.0.4 → llm_bridge-2.0.6}/llm_bridge/client/implementations/claude/stream_claude_client.py

@@ -25,6 +25,7 @@ class StreamClaudeClient(ClaudeClient):
                     messages=serialize(self.messages),
                     betas=self.betas,
                     tools=self.tools,
+                    cache_control=self.cache_control,
                     thinking=self.thinking,
                     output_config=self.output_config,
                 ) as stream:

{llm_bridge-2.0.4 → llm_bridge-2.0.6}/llm_bridge/client/model_client/claude_client.py

@@ -3,7 +3,8 @@ from typing import AsyncGenerator
 import anthropic
 from anthropic import Omit
 from anthropic.types import AnthropicBetaParam
-from anthropic.types.beta import BetaToolUnionParam, BetaOutputConfigParam, BetaThinkingConfigParam
+from anthropic.types.beta import BetaToolUnionParam, BetaOutputConfigParam, BetaThinkingConfigParam, \
+    BetaCacheControlEphemeralParam
 
 from llm_bridge.client.chat_client import ChatClient
 from llm_bridge.type.chat_response import ChatResponse
@@ -21,6 +22,7 @@ class ClaudeClient(ChatClient):
             max_tokens: int,
             betas: list[AnthropicBetaParam],
             tools: list[BetaToolUnionParam],
+            cache_control: BetaCacheControlEphemeralParam,
             thinking: BetaThinkingConfigParam | Omit,
             output_config: BetaOutputConfigParam | Omit,
     ):
@@ -32,6 +34,7 @@ class ClaudeClient(ChatClient):
         self.max_tokens = max_tokens
         self.betas = betas
         self.tools = tools
+        self.cache_control = cache_control
         self.thinking = thinking
         self.output_config = output_config
 

{llm_bridge-2.0.4 → llm_bridge-2.0.6}/llm_bridge/logic/chat_generate/model_client_factory/claude_client_factory.py

@@ -3,9 +3,10 @@ from typing import Any
 import anthropic
 from anthropic import Omit, transform_schema
 from anthropic.types import AnthropicBetaParam
-from anthropic.types.beta import BetaWebSearchTool20250305Param, BetaToolUnionParam, BetaCodeExecutionTool20250825Param, \
+from anthropic.types.beta import BetaWebSearchTool20250305Param, BetaToolUnionParam, \
     BetaJSONOutputFormatParam, BetaOutputConfigParam, BetaThinkingConfigParam, BetaThinkingConfigEnabledParam, \
-    BetaThinkingConfigAdaptiveParam
+    BetaThinkingConfigAdaptiveParam, BetaWebFetchTool20260209Param, BetaCitationsConfigParam, \
+    BetaCacheControlEphemeralParam
 
 from llm_bridge.client.implementations.claude.claude_token_counter import count_claude_input_tokens
 from llm_bridge.client.implementations.claude.non_stream_claude_client import NonStreamClaudeClient
@@ -54,6 +55,8 @@ async def create_claude_client(
         context_window - input_tokens,
     )
 
+    cache_control = BetaCacheControlEphemeralParam(type="ephemeral")
+
     thinking: BetaThinkingConfigParam | Omit = omit
     if thought:
         if model in ["claude-opus-4-7", "claude-opus-4-6", "claude-sonnet-4-6"]:
@@ -83,9 +86,11 @@ async def create_claude_client(
         )
     if code_execution:
         tools.append(
-            BetaCodeExecutionTool20250825Param(
-                type="code_execution_20250825",
-                name="code_execution",
+            # Code Execution auto-injected
+            BetaWebFetchTool20260209Param(
+                type="web_fetch_20260209",
+                name="web_fetch",
+                citations=BetaCitationsConfigParam(enabled=True)
             )
         )
 
@@ -108,6 +113,7 @@ async def create_claude_client(
             max_tokens=max_tokens,
             betas=betas,
             tools=tools,
+            cache_control=cache_control,
             thinking=thinking,
             output_config=output_config,
         )
@@ -121,6 +127,7 @@ async def create_claude_client(
             max_tokens=max_tokens,
             betas=betas,
             tools=tools,
+            cache_control=cache_control,
             thinking=thinking,
             output_config=output_config,
         )

{llm_bridge-2.0.4 → llm_bridge-2.0.6}/llm_bridge/resources/model_prices.json

@@ -1,15 +1,9 @@
 [
   {
     "apiType": "Grok",
-    "model": "grok-4.20-reasoning",
-    "input": 0.4,
-    "output": 25
-  },
-  {
-    "apiType": "Grok",
-    "model": "grok-4.20-non-reasoning",
-    "input": 0.4,
-    "output": 25
+    "model": "grok-4.3",
+    "input": 1.25,
+    "output": 2.5
   },
   {
     "apiType": "OpenAI",

{llm_bridge-2.0.4 → llm_bridge-2.0.6}/pyproject.toml

@@ -4,7 +4,7 @@ build-backend = "hatchling.build"
 
 [project]
 name = "LLM-Bridge"
-version = "2.0.4"
+version = "2.0.6"
 dependencies = [
     "fastapi",
     "httpx",

{llm_bridge-2.0.4 → llm_bridge-2.0.6}/usage/single/config.py

@@ -131,10 +131,10 @@ messages = [
             # Content(type=ContentType.Text, data="What's the weather in NYC today?"),
 
             # URL Context
-            # Content(type=ContentType.Text, data="What is in https://www.windsnow1025.com/"),
+            Content(type=ContentType.Text, data="What is in https://www.windsnow1025.com/"),
 
             # Code Execution
-            Content(type=ContentType.Text, data="What is the sum of the first 50 prime numbers? Generate and run code for the calculation, and make sure you get all 50."),
+            # Content(type=ContentType.Text, data="What is the sum of the first 50 prime numbers? Generate and run code for the calculation, and make sure you get all 50."),
 
             # Structured Output
             # Content(type=ContentType.Text, data="Please generate a product."),
@@ -162,29 +162,24 @@ messages = [
 
 # See /llm_bridge/resources/model_prices.json for available models
 # model = "gpt-5.4-mini"
-# model = "gpt-5.2"
-# model = "gpt-5.1"
 # model = "gpt-5.4"
-model = "gpt-5.4-pro"
-# model = "gpt-5"
+# model = "gpt-5.4-pro"
 # model = "gpt-4.1"
-# model = "gpt-5-codex"
-# model = "gemini-3-pro-preview"
+# model = "gemini-3.1-pro-preview"
 # model = "gemini-3-pro-image-preview"
 # model = "gemini-3.1-flash-image-preview"
 # model = "gemini-3-flash-preview"
 # model = "claude-sonnet-4-6"
-# model = "grok-4.20-reasoning"
-# model = "grok-4.20-non-reasoning"
+model = "grok-4.3"
 
 # api_type = "Vertex AI"
 # api_type = "Google AI Studio Free Tier"
 # api_type = "Google AI Studio"
-api_type = "OpenAI"
+# api_type = "OpenAI"
 # api_type = "OpenAI-Azure"
 # api_type = "OpenAI-GitHub"
 # api_type = "Claude"
-# api_type = "Grok"
+api_type = "Grok"
 
 temperature = 0
 stream = True

{llm_bridge-2.0.4 → llm_bridge-2.0.6}/uv.lock

@@ -748,7 +748,7 @@ wheels = [
 
 [[package]]
 name = "llm-bridge"
-version = "2.0.4"
+version = "2.0.6"
 source = { editable = "." }
 dependencies = [
     { name = "anthropic" },