kailash 0.1.2__py3-none-any.whl → 0.1.4__py3-none-any.whl

kailash/tracking/storage/filesystem.py

@@ -270,7 +270,6 @@ class FileSystemStorage(StorageBackend):
         run = WorkflowRun.model_validate(run_data)
 
         # Generate new run ID to avoid conflicts
-        original_run_id = run.run_id
         run.run_id = str(uuid4())
 
         # Save run

kailash/utils/export.py

@@ -88,8 +88,8 @@ class NodeMapper:
             resources=ResourceSpec(cpu="100m", memory="256Mi"),
         )
 
-        self.mappings["CSVReader"] = ContainerMapping(
-            python_node="CSVReader",
+        self.mappings["CSVReaderNode"] = ContainerMapping(
+            python_node="CSVReaderNode",
             container_image="kailash/csv-reader:latest",
             command=["python", "-m", "kailash.nodes.data.csv_reader"],
             resources=ResourceSpec(cpu="100m", memory="512Mi"),

kailash/utils/templates.py

@@ -352,7 +352,7 @@ See `workflows/example_workflow.py` for a basic workflow example.
         # Create example workflow
         workflow_content = '''"""Example workflow for data processing."""
 from kailash.workflow import Workflow
-from kailash.nodes.data import CSVReader, CSVWriter
+from kailash.nodes.data import CSVReaderNode, CSVWriterNode
 from kailash.nodes.transform import Filter, Sort
 from kailash.nodes.logic import Aggregator
 
@@ -363,11 +363,11 @@ workflow = Workflow(
 )
 
 # Add nodes
-workflow.add_node("reader", CSVReader(), file_path="examples/examples/data/input.csv")
+workflow.add_node("reader", CSVReaderNode(), file_path="examples/examples/data/input.csv")
 workflow.add_node("filter", Filter(), field="value", operator=">", value=100)
 workflow.add_node("sort", Sort(), field="value", reverse=True)
 workflow.add_node("aggregate", Aggregator(), group_by="category", operation="sum")
-workflow.add_node("writer", CSVWriter(), file_path="outputs/results.csv")
+workflow.add_node("writer", CSVWriterNode(), file_path="outputs/results.csv")
 
 # Connect nodes
 workflow.connect("reader", "filter", {"data": "data"})
@@ -471,9 +471,9 @@ Thumbs.db
         # Add data processing workflow
         workflow_content = '''"""Data processing pipeline workflow."""
 from kailash.workflow import Workflow
-from kailash.nodes.data import CSVReader, JSONReader, JSONWriter
+from kailash.nodes.data import CSVReaderNode, JSONReaderNode, JSONWriterNode
 from kailash.nodes.transform import Filter, Map, Sort
-from kailash.nodes.logic import Aggregator, Merge
+from kailash.nodes.logic import Aggregator, MergeNode
 
 # Create workflow
 workflow = Workflow(
@@ -482,20 +482,20 @@ workflow = Workflow(
 )
 
 # Data ingestion
-workflow.add_node("csv_reader", CSVReader(), file_path="examples/examples/data/sales_data.csv")
-workflow.add_node("json_reader", JSONReader(), file_path="examples/examples/data/product_data.json")
+workflow.add_node("csv_reader", CSVReaderNode(), file_path="examples/examples/data/sales_data.csv")
+workflow.add_node("json_reader", JSONReaderNode(), file_path="examples/examples/data/product_data.json")
 
 # Transform data
 workflow.add_node("filter_sales", Filter(), field="amount", operator=">", value=1000)
 workflow.add_node("calculate_profit", Map(), field="amount", operation="multiply", value=0.2)
-workflow.add_node("merge_data", Merge(), merge_type="merge_dict", key="product_id")
+workflow.add_node("merge_data", MergeNode(), merge_type="merge_dict", key="product_id")
 
 # Aggregate results
 workflow.add_node("group_by_category", Aggregator(), group_by="category", operation="sum")
 workflow.add_node("sort_results", Sort(), field="value", reverse=True)
 
 # Export results
-workflow.add_node("write_json", JSONWriter(), file_path="outputs/analysis_results.json")
+workflow.add_node("write_json", JSONWriterNode(), file_path="outputs/analysis_results.json")
 
 # Connect pipeline
 workflow.connect("csv_reader", "filter_sales", {"data": "data"})
@@ -537,7 +537,7 @@ workflow.connect("sort_results", "write_json", {"sorted_data": "data"})
         # Add ML workflow
         workflow_content = '''"""Machine learning pipeline workflow."""
 from kailash.workflow import Workflow
-from kailash.nodes.data import CSVReader, JSONWriter
+from kailash.nodes.data import CSVReaderNode, JSONWriterNode
 from kailash.nodes.transform import Filter, Map
 from kailash.nodes.ai import (
     TextClassifier,
@@ -553,7 +553,7 @@ workflow = Workflow(
 )
 
 # Data ingestion
-workflow.add_node("read_data", CSVReader(), file_path="examples/examples/data/text_data.csv")
+workflow.add_node("read_data", CSVReaderNode(), file_path="examples/examples/data/text_data.csv")
 
 # Preprocessing
 workflow.add_node("extract_text", Map(), field="content")
@@ -567,10 +567,10 @@ workflow.add_node("extract_entities", NamedEntityRecognizer(),
 workflow.add_node("summarize", TextSummarizer(), max_length=100)
 
 # Combine results
-workflow.add_node("merge_results", Merge(), merge_type="merge_dict")
+workflow.add_node("merge_results", MergeNode(), merge_type="merge_dict")
 
 # Export results
-workflow.add_node("save_results", JSONWriter(), file_path="outputs/ml_results.json")
+workflow.add_node("save_results", JSONWriterNode(), file_path="outputs/ml_results.json")
 
 # Connect pipeline
 workflow.connect("read_data", "extract_text", {"data": "data"})
@@ -604,7 +604,7 @@ workflow.connect("merge_results", "save_results", {"merged_data": "data"})
         # Add API workflow
         workflow_content = '''"""API integration workflow."""
 from kailash.workflow import Workflow
-from kailash.nodes.data import JSONReader, JSONWriter
+from kailash.nodes.data import JSONReaderNode, JSONWriterNode
 from kailash.nodes.transform import Map, Filter
 from kailash.nodes.logic import Conditional
 from kailash.nodes.ai import ChatAgent, FunctionCallingAgent
@@ -616,7 +616,7 @@ workflow = Workflow(
 )
 
 # Read configuration
-workflow.add_node("read_config", JSONReader(), file_path="examples/examples/data/api_config.json")
+workflow.add_node("read_config", JSONReaderNode(), file_path="examples/examples/data/api_config.json")
 
 # Process with AI agent
 workflow.add_node("chat_agent", ChatAgent(),
@@ -644,7 +644,7 @@ workflow.add_node("process_success", Map(), operation="identity")
 workflow.add_node("handle_error", Map(), operation="identity")
 
 # Save results
-workflow.add_node("save_results", JSONWriter(), file_path="outputs/api_results.json")
+workflow.add_node("save_results", JSONWriterNode(), file_path="outputs/api_results.json")
 
 # Connect workflow
 workflow.connect("read_config", "chat_agent", {"data": "messages"})

kailash/visualization/performance.py

@@ -136,7 +136,7 @@ class PerformanceVisualizer:
 
         # Calculate timeline bounds
         min_time = min(t.started_at for t in tasks_with_times)
-        max_time = max(t.ended_at for t in tasks_with_times)
+        max(t.ended_at for t in tasks_with_times)
 
         # Create timeline bars
         y_positions = []
@@ -266,8 +266,8 @@ class PerformanceVisualizer:
                 )
 
         # Memory usage chart
-        bars2 = ax2.bar(x, memory_usage, color="lightgreen", edgecolor="black")
-        bars2_delta = ax2.bar(
+        ax2.bar(x, memory_usage, color="lightgreen", edgecolor="black")
+        ax2.bar(
             x,
             memory_delta,
             bottom=memory_usage,
@@ -482,7 +482,7 @@ class PerformanceVisualizer:
         width = 0.35
 
         # I/O bytes chart
-        bars1 = ax1.bar(
+        ax1.bar(
             x - width / 2,
             io_read_bytes,
             width,
@@ -490,7 +490,7 @@ class PerformanceVisualizer:
             color="lightblue",
             edgecolor="black",
         )
-        bars2 = ax1.bar(
+        ax1.bar(
             x + width / 2,
             io_write_bytes,
             width,
@@ -507,7 +507,7 @@ class PerformanceVisualizer:
         ax1.grid(True, axis="y", alpha=0.3)
 
         # I/O operations count chart
-        bars3 = ax2.bar(
+        ax2.bar(
             x - width / 2,
             io_read_count,
             width,
@@ -515,7 +515,7 @@ class PerformanceVisualizer:
             color="lightblue",
             edgecolor="black",
         )
-        bars4 = ax2.bar(
+        ax2.bar(
             x + width / 2,
             io_write_count,
             width,

kailash/visualization/reports.py

@@ -639,7 +639,7 @@ class WorkflowPerformanceReporter:
 
         summary = analysis["summary"]
         bottlenecks = analysis["bottlenecks"]
-        resource_analysis = analysis["resource_analysis"]
+        analysis["resource_analysis"]
         error_analysis = analysis["error_analysis"]
 
         # Efficiency insights

kailash/workflow/graph.py

@@ -10,11 +10,11 @@ import networkx as nx
 import yaml
 from pydantic import BaseModel, Field, ValidationError
 
-from kailash.nodes import Node
+from kailash.nodes.base import Node
 
 try:
     # For normal runtime, use the actual registry
-    from kailash.nodes import NodeRegistry
+    from kailash.nodes.base import NodeRegistry
 except ImportError:
     # For tests, use the mock registry
     from kailash.workflow.mock_registry import MockRegistry as NodeRegistry
@@ -815,8 +815,8 @@ class Workflow:
 
                 # Try to find another key with a BaseModel
                 for key, value in last_node_results.items():
-                    if isinstance(value, BaseModel) and type(value) == type(
-                        state_model
+                    if isinstance(value, BaseModel) and isinstance(
+                        value, type(state_model)
                     ):
                         return value, results
 

kailash/workflow/mock_registry.py

@@ -43,7 +43,7 @@ NODE_TYPES = [
 for node_type in NODE_TYPES:
     try:
         NodeRegistry._registry[node_type] = MockNode
-    except:
+    except Exception:
         pass
 
 

{kailash-0.1.2.dist-info → kailash-0.1.4.dist-info}/METADATA

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: kailash
-Version: 0.1.2
+Version: 0.1.4
 Summary: Python SDK for the Kailash container-node architecture
 Home-page: https://github.com/integrum/kailash-python-sdk
 Author: Integrum
@@ -41,7 +41,8 @@ Requires-Dist: autodoc>=0.5.0
 Requires-Dist: myst-parser>=4.0.1
 Requires-Dist: black>=25.1.0
 Requires-Dist: psutil>=7.0.0
-Requires-Dist: fastapi[all]>=0.115.12
+Requires-Dist: fastapi>=0.115.12
+Requires-Dist: uvicorn[standard]>=0.31.0
 Requires-Dist: pytest-asyncio>=1.0.0
 Requires-Dist: pre-commit>=4.2.0
 Requires-Dist: twine>=6.1.0
@@ -65,7 +66,7 @@ Dynamic: requires-python
   <a href="https://pepy.tech/project/kailash"><img src="https://static.pepy.tech/badge/kailash" alt="Downloads"></a>
   <img src="https://img.shields.io/badge/license-MIT-green.svg" alt="MIT License">
   <img src="https://img.shields.io/badge/code%20style-black-000000.svg" alt="Code style: black">
-  <img src="https://img.shields.io/badge/tests-746%20passing-brightgreen.svg" alt="Tests: 746 passing">
+  <img src="https://img.shields.io/badge/tests-753%20passing-brightgreen.svg" alt="Tests: 753 passing">
   <img src="https://img.shields.io/badge/coverage-100%25-brightgreen.svg" alt="Coverage: 100%">
 </p>
 
@@ -89,6 +90,8 @@ Dynamic: requires-python
 - ⚡ **Fast Installation**: Uses `uv` for lightning-fast Python package management
 - 🤖 **AI-Powered**: Complete LLM agents, embeddings, and hierarchical RAG architecture
 - 🧠 **Retrieval-Augmented Generation**: Full RAG pipeline with intelligent document processing
+- 🌐 **REST API Wrapper**: Expose any workflow as a production-ready API in 3 lines
+- 🚪 **Multi-Workflow Gateway**: Manage multiple workflows through unified API with MCP integration
 
 ## 🎯 Who Is This For?
 
@@ -122,7 +125,7 @@ uv sync
 
 ```python
 from kailash.workflow import Workflow
-from kailash.nodes.data import CSVReader
+from kailash.nodes.data import CSVReaderNode
 from kailash.nodes.code import PythonCodeNode
 from kailash.runtime.local import LocalRuntime
 import pandas as pd
@@ -131,7 +134,7 @@ import pandas as pd
 workflow = Workflow("customer_analysis", name="customer_analysis")
 
 # Add data reader
-reader = CSVReader(file_path="customers.csv")
+reader = CSVReaderNode(file_path="customers.csv")
 workflow.add_node("read_customers", reader)
 
 # Add custom processing using Python code
@@ -169,7 +172,7 @@ workflow.save("customer_analysis.yaml", format="yaml")
 
 ```python
 from kailash.workflow import Workflow
-from kailash.nodes.data import SharePointGraphReader, CSVWriter
+from kailash.nodes.data import SharePointGraphReader, CSVWriterNode
 import os
 
 # Create workflow for SharePoint file processing
@@ -180,7 +183,7 @@ sharepoint = SharePointGraphReader()
 workflow.add_node("read_sharepoint", sharepoint)
 
 # Process downloaded files
-csv_writer = CSVWriter(file_path="sharepoint_output.csv")
+csv_writer = CSVWriterNode(file_path="sharepoint_output.csv")
 workflow.add_node("save_locally", csv_writer)
 
 # Connect nodes
@@ -208,8 +211,8 @@ results, run_id = runtime.execute(workflow, inputs=inputs)
 
 ```python
 from kailash.workflow import Workflow
-from kailash.nodes.ai.embedding_generator import EmbeddingGenerator
-from kailash.nodes.ai.llm_agent import LLMAgent
+from kailash.nodes.ai.embedding_generator import EmbeddingGeneratorNode
+from kailash.nodes.ai.llm_agent import LLMAgentNode
 from kailash.nodes.data.sources import DocumentSourceNode, QuerySourceNode
 from kailash.nodes.data.retrieval import RelevanceScorerNode
 from kailash.nodes.transform.chunkers import HierarchicalChunkerNode
@@ -230,17 +233,17 @@ chunk_text_extractor = ChunkTextExtractorNode()
 query_text_wrapper = QueryTextWrapperNode()
 
 # AI processing with Ollama
-chunk_embedder = EmbeddingGenerator(
+chunk_embedder = EmbeddingGeneratorNode(
     provider="ollama", model="nomic-embed-text", operation="embed_batch"
 )
-query_embedder = EmbeddingGenerator(
+query_embedder = EmbeddingGeneratorNode(
     provider="ollama", model="nomic-embed-text", operation="embed_batch"
 )
 
 # Retrieval and response generation
 relevance_scorer = RelevanceScorerNode()
 context_formatter = ContextFormatterNode()
-llm_agent = LLMAgent(provider="ollama", model="llama3.2", temperature=0.7)
+llm_agent = LLMAgentNode(provider="ollama", model="llama3.2", temperature=0.7)
 
 # Add all nodes to workflow
 for name, node in {
@@ -273,6 +276,140 @@ results, run_id = runtime.execute(workflow)
 print("RAG Response:", results["llm_agent"]["response"])
 ```
 
+### Workflow API Wrapper - Expose Workflows as REST APIs
+
+Transform any Kailash workflow into a production-ready REST API in just 3 lines of code:
+
+```python
+from kailash.api.workflow_api import WorkflowAPI
+
+# Take any workflow and expose it as an API
+api = WorkflowAPI(workflow)
+api.run(port=8000)  # That's it! Your workflow is now a REST API
+```
+
+#### Features
+
+- **Automatic REST Endpoints**:
+  - `POST /execute` - Execute workflow with inputs
+  - `GET /workflow/info` - Get workflow metadata
+  - `GET /health` - Health check endpoint
+  - Automatic OpenAPI docs at `/docs`
+
+- **Multiple Execution Modes**:
+  ```python
+  # Synchronous execution (wait for results)
+  curl -X POST http://localhost:8000/execute \
+    -d '{"inputs": {...}, "mode": "sync"}'
+
+  # Asynchronous execution (get execution ID)
+  curl -X POST http://localhost:8000/execute \
+    -d '{"inputs": {...}, "mode": "async"}'
+
+  # Check async status
+  curl http://localhost:8000/status/{execution_id}
+  ```
+
+- **Specialized APIs** for specific domains:
+  ```python
+  from kailash.api.workflow_api import create_workflow_api
+
+  # Create a RAG-specific API with custom endpoints
+  api = create_workflow_api(rag_workflow, api_type="rag")
+  # Adds /documents and /query endpoints
+  ```
+
+- **Production Ready**:
+  ```python
+  # Development
+  api.run(reload=True, log_level="debug")
+
+  # Production with SSL
+  api.run(
+      host="0.0.0.0",
+      port=443,
+      ssl_keyfile="key.pem",
+      ssl_certfile="cert.pem",
+      workers=4
+  )
+  ```
+
+See the [API demo example](examples/integration_examples/integration_api_demo.py) for complete usage patterns.
+
+### Multi-Workflow API Gateway - Manage Multiple Workflows
+
+Run multiple workflows through a single unified API gateway with dynamic routing and MCP integration:
+
+```python
+from kailash.api.gateway import WorkflowAPIGateway
+from kailash.api.mcp_integration import MCPIntegration
+
+# Create gateway
+gateway = WorkflowAPIGateway(
+    title="Enterprise Platform",
+    description="Unified API for all workflows"
+)
+
+# Register multiple workflows
+gateway.register_workflow("sales", sales_workflow)
+gateway.register_workflow("analytics", analytics_workflow)
+gateway.register_workflow("reports", reporting_workflow)
+
+# Add AI-powered tools via MCP
+mcp = MCPIntegration("ai_tools")
+mcp.add_tool("analyze", analyze_function)
+mcp.add_tool("predict", predict_function)
+gateway.register_mcp_server("ai", mcp)
+
+# Run unified server
+gateway.run(port=8000)
+```
+
+#### Gateway Features
+
+- **Unified Access Point**: All workflows accessible through one server
+  - `/sales/execute` - Execute sales workflow
+  - `/analytics/execute` - Execute analytics workflow
+  - `/workflows` - List all available workflows
+  - `/health` - Check health of all services
+
+- **MCP Integration**: AI-powered tools available to all workflows
+  ```python
+  # Use MCP tools in workflows
+  from kailash.api.mcp_integration import MCPToolNode
+
+  tool_node = MCPToolNode(
+      mcp_server="ai_tools",
+      tool_name="analyze"
+  )
+  workflow.add_node("ai_analysis", tool_node)
+  ```
+
+- **Flexible Deployment Patterns**:
+  ```python
+  # Pattern 1: Single Gateway (most cases)
+  gateway.register_workflow("workflow1", wf1)
+  gateway.register_workflow("workflow2", wf2)
+
+  # Pattern 2: Hybrid (heavy workflows separate)
+  gateway.register_workflow("light", light_wf)
+  gateway.proxy_workflow("heavy", "http://gpu-service:8080")
+
+  # Pattern 3: High Availability
+  # Run multiple gateway instances behind load balancer
+
+  # Pattern 4: Kubernetes
+  # Deploy with horizontal pod autoscaling
+  ```
+
+- **Production Features**:
+  - WebSocket support for real-time updates
+  - Health monitoring across all workflows
+  - Dynamic workflow registration/unregistration
+  - Built-in CORS and authentication support
+
+See the [Gateway examples](examples/integration_examples/gateway_comprehensive_demo.py) for complete implementation patterns.
+
 ## 📚 Documentation
 
 | Resource | Description |
@@ -294,14 +431,14 @@ The SDK includes a rich set of pre-built nodes for common operations:
 <td width="50%">
 
 **Data Operations**
-- `CSVReader` - Read CSV files
-- `JSONReader` - Read JSON files
+- `CSVReaderNode` - Read CSV files
+- `JSONReaderNode` - Read JSON files
 - `DocumentSourceNode` - Sample document provider
 - `QuerySourceNode` - Sample query provider
 - `RelevanceScorerNode` - Multi-method similarity
 - `SQLDatabaseNode` - Query databases
-- `CSVWriter` - Write CSV files
-- `JSONWriter` - Write JSON files
+- `CSVWriterNode` - Write CSV files
+- `JSONWriterNode` - Write JSON files
 
 </td>
 <td width="50%">
@@ -316,14 +453,19 @@ The SDK includes a rich set of pre-built nodes for common operations:
 - `Filter` - Filter records
 - `Aggregator` - Aggregate data
 
+**Logic Nodes**
+- `SwitchNode` - Conditional routing
+- `MergeNode` - Combine multiple inputs
+- `WorkflowNode` - Wrap workflows as reusable nodes
+
 </td>
 </tr>
 <tr>
 <td width="50%">
 
 **AI/ML Nodes**
-- `LLMAgent` - Multi-provider LLM with memory & tools
-- `EmbeddingGenerator` - Vector embeddings with caching
+- `LLMAgentNode` - Multi-provider LLM with memory & tools
+- `EmbeddingGeneratorNode` - Vector embeddings with caching
 - `MCPClient/MCPServer` - Model Context Protocol
 - `TextClassifier` - Text classification
 - `SentimentAnalyzer` - Sentiment analysis
@@ -363,14 +505,14 @@ The SDK includes a rich set of pre-built nodes for common operations:
 #### Workflow Management
 ```python
 from kailash.workflow import Workflow
-from kailash.nodes.logic import Switch
+from kailash.nodes.logic import SwitchNode
 from kailash.nodes.transform import DataTransformer
 
 # Create complex workflows with branching logic
 workflow = Workflow("data_pipeline", name="data_pipeline")
 
-# Add conditional branching with Switch node
-switch = Switch()
+# Add conditional branching with SwitchNode
+switch = SwitchNode()
 workflow.add_node("route", switch)
 
 # Different paths based on validation
@@ -384,6 +526,35 @@ workflow.connect("route", "process_valid")
 workflow.connect("route", "handle_errors")
 ```
 
+#### Hierarchical Workflow Composition
+```python
+from kailash.workflow import Workflow
+from kailash.nodes.logic import WorkflowNode
+from kailash.runtime.local import LocalRuntime
+
+# Create a reusable data processing workflow
+inner_workflow = Workflow("data_processor", name="Data Processor")
+# ... add nodes to inner workflow ...
+
+# Wrap the workflow as a node
+processor_node = WorkflowNode(
+    workflow=inner_workflow,
+    name="data_processor"
+)
+
+# Use in a larger workflow
+main_workflow = Workflow("main", name="Main Pipeline")
+main_workflow.add_node("process", processor_node)
+main_workflow.add_node("analyze", analyzer_node)
+
+# Connect workflows
+main_workflow.connect("process", "analyze")
+
+# Execute - parameters automatically mapped to inner workflow
+runtime = LocalRuntime()
+results, _ = runtime.execute(main_workflow)
+```
+
 #### Immutable State Management
 ```python
 from kailash.workflow import Workflow
@@ -667,13 +838,13 @@ chunk_text_extractor = ChunkTextExtractorNode()
 query_text_wrapper = QueryTextWrapperNode()
 
 # Create embedding generators
-chunk_embedder = EmbeddingGenerator(
+chunk_embedder = EmbeddingGeneratorNode(
     provider="ollama",
     model="nomic-embed-text",
     operation="embed_batch"
 )
 
-query_embedder = EmbeddingGenerator(
+query_embedder = EmbeddingGeneratorNode(
     provider="ollama",
     model="nomic-embed-text",
     operation="embed_batch"
@@ -684,7 +855,7 @@ relevance_scorer = RelevanceScorerNode(similarity_method="cosine")
 context_formatter = ContextFormatterNode()
 
 # Create LLM agent for final answer generation
-llm_agent = LLMAgent(
+llm_agent = LLMAgentNode(
     provider="ollama",
     model="llama3.2",
     temperature=0.7,
@@ -803,10 +974,10 @@ kailash/
 The SDK features a unified provider architecture for AI capabilities:
 
 ```python
-from kailash.nodes.ai import LLMAgent, EmbeddingGenerator
+from kailash.nodes.ai import LLMAgentNode, EmbeddingGeneratorNode
 
 # Multi-provider LLM support
-agent = LLMAgent()
+agent = LLMAgentNode()
 result = agent.run(
     provider="ollama",  # or "openai", "anthropic", "mock"
     model="llama3.1:8b-instruct-q8_0",
@@ -815,7 +986,7 @@ result = agent.run(
 )
 
 # Vector embeddings with the same providers
-embedder = EmbeddingGenerator()
+embedder = EmbeddingGeneratorNode()
 embedding = embedder.run(
     provider="ollama",  # Same providers support embeddings
     model="snowflake-arctic-embed2",