gllm-inference-binary 0.5.38__cp312-cp312-win_amd64.whl → 0.5.41__cp312-cp312-win_amd64.whl

gllm_inference/em_invoker/azure_openai_em_invoker.pyi

@@ -57,9 +57,9 @@ class AzureOpenAIEMInvoker(OpenAIEMInvoker):
 
         Retry config examples:
         ```python
-        retry_config = RetryConfig(max_retries=0, timeout=0.0)  # No retry, no timeout
+        retry_config = RetryConfig(max_retries=0, timeout=None)  # No retry, no timeout
         retry_config = RetryConfig(max_retries=0, timeout=10.0)  # No retry, 10.0 seconds timeout
-        retry_config = RetryConfig(max_retries=5, timeout=0.0)  # 5 max retries, no timeout
+        retry_config = RetryConfig(max_retries=5, timeout=None)  # 5 max retries, no timeout
         retry_config = RetryConfig(max_retries=5, timeout=10.0)  # 5 max retries, 10.0 seconds timeout
         ```
 

gllm_inference/em_invoker/bedrock_em_invoker.pyi

@@ -67,9 +67,9 @@ class BedrockEMInvoker(BaseEMInvoker):
 
         Retry config examples:
         ```python
-        retry_config = RetryConfig(max_retries=0, timeout=0.0)  # No retry, no timeout
+        retry_config = RetryConfig(max_retries=0, timeout=None)  # No retry, no timeout
         retry_config = RetryConfig(max_retries=0, timeout=10.0)  # No retry, 10.0 seconds timeout
-        retry_config = RetryConfig(max_retries=5, timeout=0.0)  # 5 max retries, no timeout
+        retry_config = RetryConfig(max_retries=5, timeout=None)  # 5 max retries, no timeout
         retry_config = RetryConfig(max_retries=5, timeout=10.0)  # 5 max retries, 10.0 seconds timeout
         ```
 

gllm_inference/em_invoker/google_em_invoker.pyi

@@ -89,9 +89,9 @@ class GoogleEMInvoker(BaseEMInvoker):
 
         Retry config examples:
         ```python
-        retry_config = RetryConfig(max_retries=0, timeout=0.0)  # No retry, no timeout
+        retry_config = RetryConfig(max_retries=0, timeout=None)  # No retry, no timeout
         retry_config = RetryConfig(max_retries=0, timeout=10.0)  # No retry, 10.0 seconds timeout
-        retry_config = RetryConfig(max_retries=5, timeout=0.0)  # 5 max retries, no timeout
+        retry_config = RetryConfig(max_retries=5, timeout=None)  # 5 max retries, no timeout
         retry_config = RetryConfig(max_retries=5, timeout=10.0)  # 5 max retries, 10.0 seconds timeout
         ```
 

gllm_inference/em_invoker/openai_em_invoker.pyi

@@ -85,9 +85,9 @@ class OpenAIEMInvoker(BaseEMInvoker):
 
         Retry config examples:
         ```python
-        retry_config = RetryConfig(max_retries=0, timeout=0.0)  # No retry, no timeout
+        retry_config = RetryConfig(max_retries=0, timeout=None)  # No retry, no timeout
         retry_config = RetryConfig(max_retries=0, timeout=10.0)  # No retry, 10.0 seconds timeout
-        retry_config = RetryConfig(max_retries=5, timeout=0.0)  # 5 max retries, no timeout
+        retry_config = RetryConfig(max_retries=5, timeout=None)  # 5 max retries, no timeout
         retry_config = RetryConfig(max_retries=5, timeout=10.0)  # 5 max retries, 10.0 seconds timeout
         ```
 

gllm_inference/em_invoker/twelevelabs_em_invoker.pyi

@@ -71,9 +71,9 @@ class TwelveLabsEMInvoker(BaseEMInvoker):
 
         Retry config examples:
         ```python
-        retry_config = RetryConfig(max_retries=0, timeout=0.0)  # No retry, no timeout
+        retry_config = RetryConfig(max_retries=0, timeout=None)  # No retry, no timeout
         retry_config = RetryConfig(max_retries=0, timeout=10.0)  # No retry, 10.0 seconds timeout
-        retry_config = RetryConfig(max_retries=5, timeout=0.0)  # 5 max retries, no timeout
+        retry_config = RetryConfig(max_retries=5, timeout=None)  # 5 max retries, no timeout
         retry_config = RetryConfig(max_retries=5, timeout=10.0)  # 5 max retries, 10.0 seconds timeout
         ```
 

gllm_inference/em_invoker/voyage_em_invoker.pyi

@@ -74,9 +74,9 @@ class VoyageEMInvoker(BaseEMInvoker):
 
         Retry config examples:
         ```python
-        retry_config = RetryConfig(max_retries=0, timeout=0.0)  # No retry, no timeout
+        retry_config = RetryConfig(max_retries=0, timeout=None)  # No retry, no timeout
         retry_config = RetryConfig(max_retries=0, timeout=10.0)  # No retry, 10.0 seconds timeout
-        retry_config = RetryConfig(max_retries=5, timeout=0.0)  # 5 max retries, no timeout
+        retry_config = RetryConfig(max_retries=5, timeout=None)  # 5 max retries, no timeout
         retry_config = RetryConfig(max_retries=5, timeout=10.0)  # 5 max retries, 10.0 seconds timeout
         ```
 

gllm_inference/lm_invoker/anthropic_lm_invoker.pyi

@@ -149,9 +149,9 @@ class AnthropicLMInvoker(BaseLMInvoker):
 
         Retry config examples:
         ```python
-        retry_config = RetryConfig(max_retries=0, timeout=0.0)  # No retry, no timeout
+        retry_config = RetryConfig(max_retries=0, timeout=None)  # No retry, no timeout
         retry_config = RetryConfig(max_retries=0, timeout=10.0)  # No retry, 10.0 seconds timeout
-        retry_config = RetryConfig(max_retries=5, timeout=0.0)  # 5 max retries, no timeout
+        retry_config = RetryConfig(max_retries=5, timeout=None)  # 5 max retries, no timeout
         retry_config = RetryConfig(max_retries=5, timeout=10.0)  # 5 max retries, 10.0 seconds timeout
         ```
 
@@ -188,17 +188,18 @@ class AnthropicLMInvoker(BaseLMInvoker):
         )
         ```
 
-        When streaming is enabled, the thinking token will be streamed with the `EventType.DATA` event type.
-
         Streaming output example:
         ```python
-        {"type": "data", "value": \'{"data_type": "thinking_start", "data_value": ""}\', ...}
-        {"type": "data", "value": \'{"data_type": "thinking", "data_value": "Let me think "}\', ...}
-        {"type": "data", "value": \'{"data_type": "thinking", "data_value": "about it..."}\', ...}
-        {"type": "data", "value": \'{"data_type": "thinking_end", "data_value": ""}\', ...}
+        {"type": "thinking_start", "value": "", ...}
+        {"type": "thinking", "value": "Let me think "\', ...}
+        {"type": "thinking", "value": "about it..."}\', ...}
+        {"type": "thinking_end", "value": ""}\', ...}
         {"type": "response", "value": "Golden retriever ", ...}
         {"type": "response", "value": "is a good dog breed.", ...}
         ```
+        Note: By default, the thinking token will be streamed with the legacy `EventType.DATA` event type.
+        To use the new simplified streamed event format, set the `simplify_events` parameter to `True` during
+        LM invoker initialization. The legacy event format support will be removed in v0.6.
 
     Batch processing:
         The `AnthropicLMInvoker` supports batch processing, which allows the language model to process multiple
@@ -265,30 +266,20 @@ class AnthropicLMInvoker(BaseLMInvoker):
 
     Output types:
         The output of the `AnthropicLMInvoker` can either be:
-        1. `str`: The text response if no additional output is needed.
-        2. `LMOutput`: A Pydantic model with the following attributes if any additional output is needed:
-            2.1. response (str): The text response.
-            2.2. tool_calls (list[ToolCall]): The tool calls, if the `tools` parameter is defined and the language
-                model decides to invoke tools. Defaults to an empty list.
-            2.3. structured_output (dict[str, Any] | BaseModel | None): The structured output, if the `response_schema`
-                parameter is defined. Defaults to None.
-            2.4. token_usage (TokenUsage | None): The token usage information, if the `output_analytics` parameter is
-                set to `True`. Defaults to None.
-            2.5. duration (float | None): The duration of the invocation in seconds, if the `output_analytics`
-                parameter is set to `True`. Defaults to None.
-            2.6. finish_details (dict[str, Any]): The details about how the generation finished, if the
-                `output_analytics` parameter is set to `True`. Defaults to an empty dictionary.
-            2.7. reasoning (list[Reasoning]): The reasoning objects, if the `thinking` parameter is set to `True`.
-                Defaults to an empty list.
-            2.8. citations (list[Chunk]): The citations. Currently not supported. Defaults to an empty list.
-            2.9. code_exec_results (list[CodeExecResult]): The code execution results. Currently not supported.
-                Defaults to an empty list.
-            2.10. mcp_calls (list[MCPCall]): The MCP calls. Currently not supported. Defaults to an empty list.
+        1. `str`: A text response.
+        2. `LMOutput`: A Pydantic model that may contain the following attributes:
+            2.1. response (str)
+            2.2. tool_calls (list[ToolCall])
+            2.3. structured_output (dict[str, Any] | BaseModel | None)
+            2.4. token_usage (TokenUsage | None)
+            2.5. duration (float | None)
+            2.6. finish_details (dict[str, Any])
+            2.7. reasoning (list[Reasoning])
     '''
     client: Incomplete
     thinking: Incomplete
     thinking_budget: Incomplete
-    def __init__(self, model_name: str, api_key: str | None = None, model_kwargs: dict[str, Any] | None = None, default_hyperparameters: dict[str, Any] | None = None, tools: list[Tool | LangChainTool] | None = None, response_schema: ResponseSchema | None = None, output_analytics: bool = False, retry_config: RetryConfig | None = None, thinking: bool = False, thinking_budget: int = ...) -> None:
+    def __init__(self, model_name: str, api_key: str | None = None, model_kwargs: dict[str, Any] | None = None, default_hyperparameters: dict[str, Any] | None = None, tools: list[Tool | LangChainTool] | None = None, response_schema: ResponseSchema | None = None, output_analytics: bool = False, retry_config: RetryConfig | None = None, thinking: bool = False, thinking_budget: int = ..., simplify_events: bool = False) -> None:
         """Initializes the AnthropicLmInvoker instance.
 
         Args:
@@ -309,6 +300,9 @@ class AnthropicLMInvoker(BaseLMInvoker):
             thinking (bool, optional): Whether to enable thinking. Only allowed for thinking models. Defaults to False.
             thinking_budget (int, optional): The tokens allocated for the thinking process. Must be greater than or
                 equal to 1024. Only allowed for thinking models. Defaults to DEFAULT_THINKING_BUDGET.
+            simplify_events (bool, optional): Temporary parameter to control the streamed events format.
+                When True, uses the simplified events format. When False, uses the legacy events format for
+                backward compatibility. Will be removed in v0.6. Defaults to False.
 
         Raises:
             ValueError:

gllm_inference/lm_invoker/azure_openai_lm_invoker.pyi

@@ -152,9 +152,9 @@ class AzureOpenAILMInvoker(OpenAILMInvoker):
 
         Retry config examples:
         ```python
-        retry_config = RetryConfig(max_retries=0, timeout=0.0)  # No retry, no timeout
+        retry_config = RetryConfig(max_retries=0, timeout=None)  # No retry, no timeout
         retry_config = RetryConfig(max_retries=0, timeout=10.0)  # No retry, 10.0 seconds timeout
-        retry_config = RetryConfig(max_retries=5, timeout=0.0)  # 5 max retries, no timeout
+        retry_config = RetryConfig(max_retries=5, timeout=None)  # 5 max retries, no timeout
         retry_config = RetryConfig(max_retries=5, timeout=10.0)  # 5 max retries, 10.0 seconds timeout
         ```
 
@@ -191,43 +191,35 @@ class AzureOpenAILMInvoker(OpenAILMInvoker):
         )
         ```
 
-        When streaming is enabled along with reasoning summary, the reasoning summary token will be streamed with the
-        `EventType.DATA` event type.
-
         Streaming output example:
         ```python
-        {"type": "data", "value": "Let me think ", ...}  # Reasoning summary token
-        {"type": "data", "value": "about it...", ...}  # Reasoning summary token
-        {"type": "response", "value": "Golden retriever ", ...}  # Response token
-        {"type": "response", "value": "is a good dog breed.", ...}  # Response token
+        {"type": "thinking_start", "value": ""}\', ...}
+        {"type": "thinking", "value": "Let me think "}\', ...}
+        {"type": "thinking", "value": "about it..."}\', ...}
+        {"type": "thinking_end", "value": ""}\', ...}
+        {"type": "response", "value": "Golden retriever ", ...}
+        {"type": "response", "value": "is a good dog breed.", ...}
         ```
+        Note: By default, the thinking token will be streamed with the legacy `EventType.DATA` event type.
+        To use the new simplified streamed event format, set the `simplify_events` parameter to `True` during
+        LM invoker initialization. The legacy event format support will be removed in v0.6.
 
         Setting reasoning-related parameters for non-reasoning models will raise an error.
 
     Output types:
         The output of the `AzureOpenAILMInvoker` can either be:
-        1. `str`: The text response if no additional output is needed.
-        2. `LMOutput`: A Pydantic model with the following attributes if any additional output is needed:
-            2.1. response (str): The text response.
-            2.2. tool_calls (list[ToolCall]): The tool calls, if the `tools` parameter is defined and the language
-                model decides to invoke tools. Defaults to an empty list.
-            2.3. structured_output (dict[str, Any] | BaseModel | None): The structured output, if the `response_schema`
-                parameter is defined. Defaults to None.
-            2.4. token_usage (TokenUsage | None): The token usage analytics, if the `output_analytics` parameter is
-                set to `True`. Defaults to None.
-            2.5. duration (float | None): The duration of the invocation in seconds, if the `output_analytics`
-                parameter is set to `True`. Defaults to None.
-            2.6. finish_details (dict[str, Any] | None): The details about how the generation finished, if the
-                `output_analytics` parameter is set to `True`. Defaults to None.
-            2.7. reasoning (list[Reasoning]): The reasoning objects, if the `reasoning_summary` parameter is provided
-                for reasoning models. Defaults to an empty list.
-            2.8. citations (list[Chunk]): The citations. Currently not supported. Defaults to an empty list.
-            2.9. code_exec_results (list[CodeExecResult]): The code execution results. Currently not supported.
-                Defaults to an empty list.
-            2.10. mcp_calls (list[MCPCall]): The MCP calls. Currently not supported. Defaults to an empty list.
+        1. `str`: A text response.
+        2. `LMOutput`: A Pydantic model that may contain the following attributes:
+            2.1. response (str)
+            2.2. tool_calls (list[ToolCall])
+            2.3. structured_output (dict[str, Any] | BaseModel | None)
+            2.4. token_usage (TokenUsage | None)
+            2.5. duration (float | None)
+            2.6. finish_details (dict[str, Any] | None)
+            2.7. reasoning (list[Reasoning])
     '''
     client_kwargs: Incomplete
-    def __init__(self, azure_endpoint: str, azure_deployment: str, api_key: str | None = None, api_version: str | None = None, model_kwargs: dict[str, Any] | None = None, default_hyperparameters: dict[str, Any] | None = None, tools: list[Tool | LangChainTool] | None = None, response_schema: ResponseSchema | None = None, output_analytics: bool = False, retry_config: RetryConfig | None = None, reasoning_effort: ReasoningEffort | None = None, reasoning_summary: ReasoningSummary | None = None) -> None:
+    def __init__(self, azure_endpoint: str, azure_deployment: str, api_key: str | None = None, api_version: str | None = None, model_kwargs: dict[str, Any] | None = None, default_hyperparameters: dict[str, Any] | None = None, tools: list[Tool | LangChainTool] | None = None, response_schema: ResponseSchema | None = None, output_analytics: bool = False, retry_config: RetryConfig | None = None, reasoning_effort: ReasoningEffort | None = None, reasoning_summary: ReasoningSummary | None = None, simplify_events: bool = False) -> None:
         """Initializes a new instance of the AzureOpenAILMInvoker class.
 
         Args:
@@ -251,6 +243,9 @@ class AzureOpenAILMInvoker(OpenAILMInvoker):
                 for non-reasoning models. If None, the model will perform medium reasoning effort. Defaults to None.
             reasoning_summary (ReasoningSummary | None, optional): The reasoning summary level for reasoning models.
                 Not allowed for non-reasoning models. If None, no summary will be generated. Defaults to None.
+            simplify_events (bool, optional): Temporary parameter to control the streamed events format.
+                When True, uses the simplified events format. When False, uses the legacy events format for
+                backward compatibility. Will be removed in v0.6. Defaults to False.
 
         Raises:
             ValueError:

gllm_inference/lm_invoker/bedrock_lm_invoker.pyi

@@ -149,9 +149,9 @@ class BedrockLMInvoker(BaseLMInvoker):
 
         Retry config examples:
         ```python
-        retry_config = RetryConfig(max_retries=0, timeout=0.0)  # No retry, no timeout
+        retry_config = RetryConfig(max_retries=0, timeout=None)  # No retry, no timeout
         retry_config = RetryConfig(max_retries=0, timeout=10.0)  # No retry, 10.0 seconds timeout
-        retry_config = RetryConfig(max_retries=5, timeout=0.0)  # 5 max retries, no timeout
+        retry_config = RetryConfig(max_retries=5, timeout=None)  # 5 max retries, no timeout
         retry_config = RetryConfig(max_retries=5, timeout=10.0)  # 5 max retries, 10.0 seconds timeout
         ```
 
@@ -162,24 +162,14 @@ class BedrockLMInvoker(BaseLMInvoker):
 
     Output types:
         The output of the `BedrockLMInvoker` can either be:
-        1. `str`: The text response if no additional output is needed.
-        2. `LMOutput`: A Pydantic model with the following attributes if any additional output is needed:
-            2.1. response (str): The text response.
-            2.2. tool_calls (list[ToolCall]): The tool calls, if the `tools` parameter is defined and the language
-                model decides to invoke tools. Defaults to an empty list.
-            2.3. structured_output (dict[str, Any] | BaseModel | None): The structured output, if the `response_schema`
-                parameter is defined. Defaults to None.
-            2.4. token_usage (TokenUsage | None): The token usage information, if the `output_analytics` parameter is
-                set to `True`. Defaults to None.
-            2.5. duration (float | None): The duration of the invocation in seconds, if the `output_analytics`
-                parameter is set to `True`. Defaults to None.
-            2.6. finish_details (dict[str, Any]): The details about how the generation finished, if the
-                `output_analytics` parameter is set to `True`. Defaults to an empty dictionary.
-            2.7. reasoning (list[Reasoning]): The reasoning objects. Currently not supported. Defaults to an empty list.
-            2.8. citations (list[Chunk]): The citations. Currently not supported. Defaults to an empty list.
-            2.9. code_exec_results (list[CodeExecResult]): The code execution results. Currently not supported.
-                Defaults to an empty list.
-            2.10. mcp_calls (list[MCPCall]): The MCP calls. Currently not supported. Defaults to an empty list.
+        1. `str`: A text response.
+        2. `LMOutput`: A Pydantic model that may contain the following attributes:
+            2.1. response (str)
+            2.2. tool_calls (list[ToolCall])
+            2.3. structured_output (dict[str, Any] | BaseModel | None)
+            2.4. token_usage (TokenUsage | None)
+            2.5. duration (float | None)
+            2.6. finish_details (dict[str, Any] | None)
     '''
     session: Incomplete
     client_kwargs: Incomplete

gllm_inference/lm_invoker/datasaur_lm_invoker.pyi

@@ -3,7 +3,7 @@ from gllm_core.event import EventEmitter as EventEmitter
 from gllm_core.schema.tool import Tool as Tool
 from gllm_core.utils.retry import RetryConfig as RetryConfig
 from gllm_inference.constants import DOCUMENT_MIME_TYPES as DOCUMENT_MIME_TYPES, INVOKER_PROPAGATED_MAX_RETRIES as INVOKER_PROPAGATED_MAX_RETRIES
-from gllm_inference.lm_invoker.openai_compatible_lm_invoker import OpenAICompatibleLMInvoker as OpenAICompatibleLMInvoker
+from gllm_inference.lm_invoker.openai_chat_completions_lm_invoker import OpenAIChatCompletionsLMInvoker as OpenAIChatCompletionsLMInvoker
 from gllm_inference.lm_invoker.schema.datasaur import InputType as InputType, Key as Key
 from gllm_inference.schema import Attachment as Attachment, AttachmentType as AttachmentType, LMOutput as LMOutput, Message as Message, ModelId as ModelId, ModelProvider as ModelProvider, ResponseSchema as ResponseSchema, ToolCall as ToolCall, ToolResult as ToolResult
 from langchain_core.tools import Tool as LangChainTool
@@ -11,7 +11,7 @@ from typing import Any
 
 SUPPORTED_ATTACHMENTS: Incomplete
 
-class DatasaurLMInvoker(OpenAICompatibleLMInvoker):
+class DatasaurLMInvoker(OpenAIChatCompletionsLMInvoker):
     '''A language model invoker to interact with Datasaur LLM Projects Deployment API.
 
     Attributes:
@@ -72,9 +72,9 @@ class DatasaurLMInvoker(OpenAICompatibleLMInvoker):
 
         Retry config examples:
         ```python
-        retry_config = RetryConfig(max_retries=0, timeout=0.0)  # No retry, no timeout
+        retry_config = RetryConfig(max_retries=0, timeout=None)  # No retry, no timeout
         retry_config = RetryConfig(max_retries=0, timeout=10.0)  # No retry, 10.0 seconds timeout
-        retry_config = RetryConfig(max_retries=5, timeout=0.0)  # 5 max retries, no timeout
+        retry_config = RetryConfig(max_retries=5, timeout=None)  # 5 max retries, no timeout
         retry_config = RetryConfig(max_retries=5, timeout=10.0)  # 5 max retries, 10.0 seconds timeout
         ```
 
@@ -103,23 +103,13 @@ class DatasaurLMInvoker(OpenAICompatibleLMInvoker):
 
     Output types:
         The output of the `DatasaurLMInvoker` can either be:
-        1. `str`: The text response if no additional output is needed.
-        2. `LMOutput`: A Pydantic model with the following attributes if any additional output is needed:
-            2.1. response (str): The text response.
-            2.2. tool_calls (list[ToolCall]): The tool calls. Currently not supported. Defaults to an empty list.
-            2.3. structured_output (dict[str, Any] | BaseModel | None): The structured output. Currently not supported.
-                Defaults to None.
-            2.4. token_usage (TokenUsage | None): The token usage analytics, if the `output_analytics` parameter is
-                set to `True`. Defaults to None.
-            2.5. duration (float | None): The duration of the invocation in seconds, if the `output_analytics`
-                parameter is set to `True`. Defaults to None.
-            2.6. finish_details (dict[str, Any] | None): The details about how the generation finished, if the
-                `output_analytics` parameter is set to `True`. Defaults to None.
-            2.7. reasoning (list[Reasoning]): The reasoning objects. Currently not supported. Defaults to an empty list.
-            2.8. citations (list[Chunk]): The citations. Currently not supported. Defaults to an empty list.
-            2.9. code_exec_results (list[CodeExecResult]): The code execution results. Currently not supported.
-                Defaults to an empty list.
-            2.10. mcp_calls (list[MCPCall]): The MCP calls. Currently not supported. Defaults to an empty list.
+        1. `str`: A text response.
+        2. `LMOutput`: A Pydantic model that may contain the following attributes:
+            2.1. response (str)
+            2.2. token_usage (TokenUsage | None)
+            2.3. duration (float | None)
+            2.4. finish_details (dict[str, Any] | None)
+            2.5. citations (list[Chunk])
     '''
     client_kwargs: Incomplete
     citations: Incomplete

gllm_inference/lm_invoker/google_lm_invoker.pyi

@@ -14,6 +14,7 @@ from typing import Any
 SUPPORTED_ATTACHMENTS: Incomplete
 DEFAULT_THINKING_BUDGET: int
 REQUIRE_THINKING_MODEL_PREFIX: Incomplete
+IMAGE_GENERATION_MODELS: Incomplete
 YOUTUBE_URL_PATTERN: Incomplete
 
 class GoogleLMInvoker(BaseLMInvoker):
@@ -30,6 +31,7 @@ class GoogleLMInvoker(BaseLMInvoker):
             structured response as defined by the schema. Supports both Pydantic BaseModel and JSON schema dictionary.
         output_analytics (bool): Whether to output the invocation analytics.
         retry_config (RetryConfig | None): The retry configuration for the language model.
+        generate_image (bool): Whether to generate image. Only allowed for image generation models.
         thinking (bool): Whether to enable thinking. Only allowed for thinking models.
         thinking_budget (int): The tokens allowed for thinking process. Only allowed for thinking models.
             If set to -1, the model will control the budget automatically.
@@ -80,6 +82,26 @@ class GoogleLMInvoker(BaseLMInvoker):
         result = await lm_invoker.invoke([text, image])
         ```
 
+    Image generation:
+        The `GoogleLMInvoker` supports image generation. This can be done by using an image generation model,
+        such as `gemini-2.5-flash-image`. Streaming is disabled for image generation models.
+        The generated image will be stored in the `attachments` attribute in the output.
+
+        Usage example:
+        ```python
+        lm_invoker = GoogleLMInvoker("gemini-2.5-flash-image")
+        result = await lm_invoker.invoke("Create a picture...")
+        result.attachments[0].write_to_file("path/to/local/image.png")
+        ```
+
+        Output example:
+        ```python
+        LMOutput(
+            response="Let me call the tools...",
+            attachments=[Attachment(filename="image.png", mime_type="image/png", data=b"...")],
+        )
+        ```
+
     Tool calling:
         Tool calling is a feature that allows the language model to call tools to perform tasks.
         Tools can be passed to the via the `tools` parameter as a list of `Tool` objects.
@@ -182,9 +204,9 @@ class GoogleLMInvoker(BaseLMInvoker):
 
         Retry config examples:
         ```python
-        retry_config = RetryConfig(max_retries=0, timeout=0.0)  # No retry, no timeout
+        retry_config = RetryConfig(max_retries=0, timeout=None)  # No retry, no timeout
         retry_config = RetryConfig(max_retries=0, timeout=10.0)  # No retry, 10.0 seconds timeout
-        retry_config = RetryConfig(max_retries=5, timeout=0.0)  # 5 max retries, no timeout
+        retry_config = RetryConfig(max_retries=5, timeout=None)  # 5 max retries, no timeout
         retry_config = RetryConfig(max_retries=5, timeout=10.0)  # 5 max retries, 10.0 seconds timeout
         ```
 
@@ -216,17 +238,18 @@ class GoogleLMInvoker(BaseLMInvoker):
         )
         ```
 
-        When streaming is enabled, the thinking token will be streamed with the `EventType.DATA` event type.
-
         Streaming output example:
         ```python
-        {"type": "data", "value": \'{"data_type": "thinking_start", "data_value": ""}\', ...}
-        {"type": "data", "value": \'{"data_type": "thinking", "data_value": "Let me think "}\', ...}
-        {"type": "data", "value": \'{"data_type": "thinking", "data_value": "about it..."}\', ...}
-        {"type": "data", "value": \'{"data_type": "thinking_end", "data_value": ""}\', ...}
+        {"type": "thinking_start", "value": "", ...}
+        {"type": "thinking", "value": "Let me think "\', ...}
+        {"type": "thinking", "value": "about it...", ...}
+        {"type": "thinking_end", "value": ""}\', ...}
         {"type": "response", "value": "Golden retriever ", ...}
         {"type": "response", "value": "is a good dog breed.", ...}
         ```
+        Note: By default, the thinking token will be streamed with the legacy `EventType.DATA` event type.
+        To use the new simplified streamed event format, set the `simplify_events` parameter to `True` during
+        LM invoker initialization. The legacy event format support will be removed in v0.6.
 
         When thinking is enabled, the amount of tokens allocated for the thinking process can be set via the
         `thinking_budget` parameter. The `thinking_budget`:
@@ -236,30 +259,22 @@ class GoogleLMInvoker(BaseLMInvoker):
 
     Output types:
         The output of the `GoogleLMInvoker` can either be:
-        1. `str`: The text response if no additional output is needed.
-        2. `LMOutput`: A Pydantic model with the following attributes if any additional output is needed:
-            2.1. response (str): The text response.
-            2.2. tool_calls (list[ToolCall]): The tool calls, if the `tools` parameter is defined and the language
-                model decides to invoke tools. Defaults to an empty list.
-            2.3. structured_output (dict[str, Any] | BaseModel | None): The structured output, if the `response_schema`
-                parameter is defined. Defaults to None.
-            2.4. token_usage (TokenUsage | None): The token usage analytics, if the `output_analytics` parameter is
-                set to `True`. Defaults to None.
-            2.5. duration (float | None): The duration of the invocation in seconds, if the `output_analytics`
-                parameter is set to `True`. Defaults to None.
-            2.6. finish_details (dict[str, Any] | None): The details about how the generation finished, if the
-                `output_analytics` parameter is set to `True`. Defaults to None.
-            2.7. reasoning (list[Reasoning]): The reasoning objects, if the `thinking` parameter is set to `True`.
-                Defaults to an empty list.
-            2.8. citations (list[Chunk]): The citations. Currently not supported. Defaults to an empty list.
-            2.9. code_exec_results (list[CodeExecResult]): The code execution results. Currently not supported.
-                Defaults to an empty list.
-            2.10. mcp_calls (list[MCPCall]): The MCP calls. Currently not supported. Defaults to an empty list.
+        1. `str`: A text response.
+        2. `LMOutput`: A Pydantic model that may contain the following attributes:
+            2.1. response (str)
+            2.2. attachments (list[Attachment])
+            2.3. tool_calls (list[ToolCall])
+            2.4. structured_output (dict[str, Any] | BaseModel | None)
+            2.5. token_usage (TokenUsage | None)
+            2.6. duration (float | None)
+            2.7. finish_details (dict[str, Any])
+            2.8. reasoning (list[Reasoning])
     '''
     client_params: Incomplete
+    generate_image: Incomplete
     thinking: Incomplete
     thinking_budget: Incomplete
-    def __init__(self, model_name: str, api_key: str | None = None, credentials_path: str | None = None, project_id: str | None = None, location: str = 'us-central1', model_kwargs: dict[str, Any] | None = None, default_hyperparameters: dict[str, Any] | None = None, tools: list[Tool | LangChainTool] | None = None, response_schema: ResponseSchema | None = None, output_analytics: bool = False, retry_config: RetryConfig | None = None, thinking: bool | None = None, thinking_budget: int = ...) -> None:
+    def __init__(self, model_name: str, api_key: str | None = None, credentials_path: str | None = None, project_id: str | None = None, location: str = 'us-central1', model_kwargs: dict[str, Any] | None = None, default_hyperparameters: dict[str, Any] | None = None, tools: list[Tool | LangChainTool] | None = None, response_schema: ResponseSchema | None = None, output_analytics: bool = False, retry_config: RetryConfig | None = None, thinking: bool | None = None, thinking_budget: int = ..., simplify_events: bool = False) -> None:
         '''Initializes a new instance of the GoogleLMInvoker class.
 
         Args:
@@ -288,6 +303,9 @@ class GoogleLMInvoker(BaseLMInvoker):
                 Defaults to True for Gemini 2.5 Pro models and False for other models.
             thinking_budget (int, optional): The tokens allowed for thinking process. Only allowed for thinking models.
                 Defaults to -1, in which case the model will control the budget automatically.
+            simplify_events (bool, optional): Temporary parameter to control the streamed events format.
+                When True, uses the simplified events format. When False, uses the legacy events format for
+                backward compatibility. Will be removed in v0.6. Defaults to False.
 
         Note:
             If neither `api_key` nor `credentials_path` is provided, Google Gen AI will be used by default.

gllm_inference/lm_invoker/langchain_lm_invoker.pyi

@@ -175,9 +175,9 @@ class LangChainLMInvoker(BaseLMInvoker):
 
         Retry config examples:
         ```python
-        retry_config = RetryConfig(max_retries=0, timeout=0.0)  # No retry, no timeout
+        retry_config = RetryConfig(max_retries=0, timeout=None)  # No retry, no timeout
         retry_config = RetryConfig(max_retries=0, timeout=10.0)  # No retry, 10.0 seconds timeout
-        retry_config = RetryConfig(max_retries=5, timeout=0.0)  # 5 max retries, no timeout
+        retry_config = RetryConfig(max_retries=5, timeout=None)  # 5 max retries, no timeout
         retry_config = RetryConfig(max_retries=5, timeout=10.0)  # 5 max retries, 10.0 seconds timeout
         ```
 
@@ -188,24 +188,14 @@ class LangChainLMInvoker(BaseLMInvoker):
 
     Output types:
         The output of the `LangChainLMInvoker` can either be:
-        1. `str`: The text response if no additional output is needed.
-        2. `LMOutput`: A Pydantic model with the following attributes if any additional output is needed:
-            2.1. response (str): The text response.
-            2.2. tool_calls (list[ToolCall]): The tool calls, if the `tools` parameter is defined and the language
-                model decides to invoke tools. Defaults to an empty list.
-            2.3. structured_output (dict[str, Any] | BaseModel | None): The structured output, if the `response_schema`
-                parameter is defined. Defaults to None.
-            2.4. token_usage (TokenUsage | None): The token usage analytics, if the `output_analytics` parameter is
-                set to `True`. Defaults to None.
-            2.5. duration (float | None): The duration of the invocation in seconds, if the `output_analytics`
-                parameter is set to `True`. Defaults to None.
-            2.6. finish_details (dict[str, Any] | None): The details about how the generation finished, if the
-                `output_analytics` parameter is set to `True`. Defaults to None.
-            2.7. reasoning (list[Reasoning]): The reasoning objects. Currently not supported. Defaults to an empty list.
-            2.8. citations (list[Chunk]): The citations. Currently not supported. Defaults to an empty list.
-            2.9. code_exec_results (list[CodeExecResult]): The code execution results. Currently not supported.
-                Defaults to an empty list.
-            2.10. mcp_calls (list[MCPCall]): The MCP calls. Currently not supported. Defaults to an empty list.
+        1. `str`: A text response.
+        2. `LMOutput`: A Pydantic model that may contain the following attributes:
+            2.1. response (str)
+            2.2. tool_calls (list[ToolCall])
+            2.3. structured_output (dict[str, Any] | BaseModel | None)
+            2.4. token_usage (TokenUsage | None)
+            2.5. duration (float | None)
+            2.6. finish_details (dict[str, Any])
     '''
     model: Incomplete
     def __init__(self, model: BaseChatModel | None = None, model_class_path: str | None = None, model_name: str | None = None, model_kwargs: dict[str, Any] | None = None, default_hyperparameters: dict[str, Any] | None = None, tools: list[Tool | LangChainTool] | None = None, response_schema: ResponseSchema | None = None, output_analytics: bool = False, retry_config: RetryConfig | None = None) -> None: