LLM nodes

Chat-completion model adapters and prompt-template helpers.

15 nodes.

---

ai_gateway_chat_model — AI Gateway Chat Model

Calls a corporate AI gateway using OpenAI-compatible, OpenAI, or Anthropic protocol.

Config fields

Field	Type	Required	Values
apiProvider	string	yes	openai_compatible | openai | anthropic
baseUrl	string	yes	—
model	string	yes	—
secretRef	object	no	—
temperature	number	no	—
maxTokens	number	no	—
enableThinking	boolean	no	—
thinkingTokens	number	no	—
promptKey	string	no	—
systemPromptKey	string	no	—

Example config

{
  "apiProvider": "openai_compatible",
  "baseUrl": "https://llm.company.example/v1",
  "model": "claude-haiku-4-5-20251001",
  "temperature": 0.2,
  "maxTokens": 1024,
  "enableThinking": false,
  "thinkingTokens": 1024,
  "promptKey": "prompt",
  "systemPromptKey": "system_prompt"
}

---

ai_transform — AI Transform

Transform data using natural language instructions. The LLM generates and applies a transformation based on your description.

Config fields

Field	Type	Required	Values
instruction	string	yes	—
inputKey	string	no	—
outputFormat	string	no	text | json | code

Example config

{
  "instruction": "Convert this data to a markdown table",
  "inputKey": "text",
  "outputFormat": "text"
}

---

anthropic_chat_model — Anthropic Chat Model

Calls Anthropic Claude models with an API key or ANTHROPIC_API_KEY.

Config fields

Field	Type	Required	Values
baseUrl	string	no	—
model	string	yes	—
secretRef	object	no	—
temperature	number	no	—
maxTokens	number	no	—
promptKey	string	no	—
systemPromptKey	string	no	—

Example config

{
  "model": "claude-3-5-sonnet-latest",
  "temperature": 0.2,
  "maxTokens": 1024,
  "promptKey": "prompt",
  "systemPromptKey": "system_prompt"
}

---

azure_openai_chat_model — Azure OpenAI Chat Model

Calls Azure OpenAI chat/completions using deployment + API version.

Config fields

Field	Type	Required	Values
endpoint	string	yes	—
deployment	string	yes	—
apiVersion	string	no	—
secretRef	object	yes	—
temperature	number	no	—
maxTokens	number	no	—
promptKey	string	no	—
systemPromptKey	string	no	—

Example config

{
  "endpoint": "https://my-azure-openai.openai.azure.com",
  "deployment": "gpt-4o-mini",
  "apiVersion": "2024-10-21",
  "temperature": 0.2,
  "maxTokens": 1024,
  "promptKey": "prompt",
  "systemPromptKey": "system_prompt"
}

---

basic_llm_chain — Basic LLM Chain

Simple prompt-to-response chain with configurable system and user prompt templates.

Config fields

Field	Type	Required	Values
promptKey	string	no	—
systemPromptKey	string	no	—
outputFormat	string	no	text | json

Example config

{
  "promptKey": "prompt",
  "systemPromptKey": "system_prompt",
  "outputFormat": "text"
}

---

google_gemini_chat_model — Google Gemini Chat Model

Calls Google Gemini chat completions using an API key.

Config fields

Field	Type	Required	Values
model	string	yes	—
secretRef	object	no	—
temperature	number	no	—
maxTokens	number	no	—
promptKey	string	no	—
systemPromptKey	string	no	—

Example config

{
  "model": "gemini-2.5-flash",
  "temperature": 0.2,
  "maxTokens": 1024,
  "promptKey": "prompt",
  "systemPromptKey": "system_prompt"
}

---

information_extractor — Information Extractor

Extracts structured information from text based on a configurable schema.

Config fields

Field	Type	Required	Values
textKey	string	no	—
extractionSchema	string	no	—

Example config

{
  "textKey": "text",
  "extractionSchema": "{ \"name\": \"string\", \"email\": \"string\", \"company\": \"string\" }"
}

---

llm_call — LLM Call

Executes a model completion call.

Config fields

Field	Type	Required	Values
provider	object	yes	—
promptKey	string	no	—
systemPromptKey	string	no	—

Example config

{
  "provider": {
    "providerId": "ollama",
    "model": "llama3.1",
    "baseUrl": "http://localhost:11434/v1"
  },
  "promptKey": "prompt",
  "systemPromptKey": "system_prompt"
}

---

ollama_chat_model — Ollama Chat Model

Calls a local Ollama model through its OpenAI-compatible API.

Config fields

Field	Type	Required	Values
baseUrl	string	no	—
model	string	yes	—
temperature	number	no	—
maxTokens	number	no	—
promptKey	string	no	—
systemPromptKey	string	no	—

Example config

{
  "baseUrl": "http://localhost:11434/v1",
  "model": "llama3.1",
  "temperature": 0.2,
  "maxTokens": 1024,
  "promptKey": "prompt",
  "systemPromptKey": "system_prompt"
}

---

openai_chat_model — OpenAI Chat Model

Calls OpenAI chat models with an API key or OPENAI_API_KEY.

Config fields

Field	Type	Required	Values
baseUrl	string	no	—
model	string	yes	—
secretRef	object	no	—
temperature	number	no	—
maxTokens	number	no	—
promptKey	string	no	—
systemPromptKey	string	no	—

Example config

{
  "model": "gpt-4o-mini",
  "temperature": 0.2,
  "maxTokens": 1024,
  "promptKey": "prompt",
  "systemPromptKey": "system_prompt"
}

---

openai_compatible_chat_model — OpenAI Compatible Chat Model

Calls any OpenAI-compatible chat API such as OpenRouter, Groq, vLLM, or LM Studio.

Config fields

Field	Type	Required	Values
baseUrl	string	yes	—
model	string	yes	—
secretRef	object	no	—
temperature	number	no	—
maxTokens	number	no	—
promptKey	string	no	—
systemPromptKey	string	no	—

Example config

{
  "baseUrl": "http://localhost:1234/v1",
  "model": "local-model",
  "temperature": 0.2,
  "maxTokens": 1024,
  "promptKey": "prompt",
  "systemPromptKey": "system_prompt"
}

---

qa_chain — Q&A Chain

Answers questions using provided context documents. Returns answer with source citations.

Config fields

Field	Type	Required	Values
contextKey	string	no	—
questionKey	string	no	—
maxContextLength	number	no	—

Example config

{
  "contextKey": "context",
  "questionKey": "question",
  "maxContextLength": 4000
}

---

sentiment_analysis — Sentiment Analysis

Analyzes sentiment of input text. Returns sentiment label, confidence score, and explanation.

Config fields

Field	Type	Required	Values
textKey	string	no	—

Example config

{
  "textKey": "text"
}

---

summarization_chain — Summarization Chain

Summarizes input text with configurable length and style.

Config fields

Field	Type	Required	Values
textKey	string	no	—
maxLength	number	no	—
style	string	no	brief | detailed | bullet_points | executive

Example config

{
  "textKey": "text",
  "maxLength": 500,
  "style": "brief"
}

---

text_classifier — Text Classifier

Classifies input text into one of the configured categories.

Config fields

Field	Type	Required	Values
textKey	string	no	—
categories	string	no	—
multiLabel	boolean	no	—

Example config

{
  "textKey": "text",
  "categories": "positive, negative, neutral",
  "multiLabel": false
}