Package com.tnsai.reasoning

Class ReasoningRouter

java.lang.Object

com.tnsai.reasoning.ReasoningRouter

All Implemented Interfaces:

com.tnsai.llm.LLMClient

public class ReasoningRouter
extends Object
implements com.tnsai.llm.LLMClient

Routes queries to reasoning-specialized models based on task complexity.

Detects reasoning tasks (math, logic, planning, coding) and routes them to specialized models like o1, DeepSeek-R1, or Qwen-Math while using standard models for simple queries.

Usage

 ReasoningRouter router = ReasoningRouter.builder()
     .reasoningModel(o1Client)      // For complex reasoning
     .standardModel(gpt4oClient)    // For simple queries
     .threshold(0.7)                // Routing threshold
     .build();

 // Automatically routes to o1 for complex math
 ChatResponse response = router.chat("Prove that sqrt(2) is irrational");

 // Uses gpt-4o for simple queries
 ChatResponse response2 = router.chat("What is the capital of France?");
 

Nested Class Summary

Nested Classes

Modifier and Type	Class	Description
static final class	ReasoningRouter.Builder	Builder for ReasoningRouter.

Method Summary

Modifier and Type	Method	Description
static ReasoningRouter.Builder	builder()	Creates a new builder for ReasoningRouter.
double	calculateReasoningScore(String query)	Calculates a reasoning score for the given query.
com.tnsai.llm.ChatResponse	chat(String message, Optional<String> systemPrompt, Optional<List<Map<String,Object>>> history, Optional<List<Map<String,Object>>> tools)
Optional<Integer>	getMaxTokens()
String	getModel()
com.tnsai.llm.LLMClient	getReasoningModel()	Gets the reasoning model.
com.tnsai.llm.LLMClient	getStandardModel()	Gets the standard model.
float	getTemperature()
double	getThreshold()	Gets the routing threshold.
com.tnsai.llm.LLMClient	selectModel(String query)	Determines which model to use for the query.
Stream<String>	streamChat(String message, Optional<String> systemPrompt, Optional<List<Map<String,Object>>> history, Optional<List<Map<String,Object>>> tools)

Methods inherited from class java.lang.Object

equals, getClass, hashCode, notify, notifyAll, toString, wait, wait, wait

Methods inherited from interface com.tnsai.llm.LLMClient

chat, chat, chat, chat, chat, getCapabilities, getProviderName, getTopP, listModels, streamChat, streamChat, streamChatWithHandler, streamChatWithHandler, streamChatWithSpec, streamChatWithSpec

Method Details

builder

public static ReasoningRouter.Builder builder()

Creates a new builder for ReasoningRouter.

calculateReasoningScore

public double calculateReasoningScore(String query)

Calculates a reasoning score for the given query.

Parameters:

query - The query to analyze

Returns:

Score between 0.0 and 1.0

selectModel

public com.tnsai.llm.LLMClient selectModel(String query)

Determines which model to use for the query.

Parameters:

query - The query to route

Returns:

The selected LLM client

getModel

public String getModel()

Specified by:

getModel in interface com.tnsai.llm.LLMClient

getTemperature

public float getTemperature()

Specified by:

getTemperature in interface com.tnsai.llm.LLMClient

getMaxTokens

public Optional<Integer> getMaxTokens()

Specified by:

getMaxTokens in interface com.tnsai.llm.LLMClient

chat

public com.tnsai.llm.ChatResponse chat(String message,
 Optional<String> systemPrompt,
 Optional<List<Map<String,Object>>> history,
 Optional<List<Map<String,Object>>> tools)

Specified by:

chat in interface com.tnsai.llm.LLMClient

streamChat

public Stream<String> streamChat(String message,
 Optional<String> systemPrompt,
 Optional<List<Map<String,Object>>> history,
 Optional<List<Map<String,Object>>> tools)

Specified by:

streamChat in interface com.tnsai.llm.LLMClient

getReasoningModel

public com.tnsai.llm.LLMClient getReasoningModel()

Gets the reasoning model.

getStandardModel

public com.tnsai.llm.LLMClient getStandardModel()

Gets the standard model.

getThreshold

public double getThreshold()

Gets the routing threshold.