v2.3.1

What's Changed

• HttpLlm.appliation() escapes some special characters. by @samchon in #117

Full Changelog: v2.3.0...v2.3.1

v2.3.0

New plugin properties OpenApi.IOperation["x-samchon-accessor"] and OpenApi.IOperation["x-samchon-controller"] for OpenAPI generator and LLM application composer, indicating operation's function name and belonged controller.

Also prohibited non-variable function name assignment.

This is a special update for commercial project performing LLM function calling to the Rest API.

What's Changed

• Test empty parameters' function calling cases by @samchon in #114
• New plugin property OpenApi.IOperation["x-samchon-accessor"]. by @samchon in #115

Full Changelog: v2.2.1...v2.3.0

v2.2.1

What's Changed

• Enhance ChatGptTypeChecker and LlmTypeCheckerV3. by @samchon in #113

Full Changelog: v2.2.0...v2.2.1

v2.2.0

As I've understood that ChatGPT's strict mode is not perfect, and found much better way by giving validation feedback with typia.validate<T>() function, I've changed the ChatGPT strict mode to be configurable, and also changed ChatGPT function calling (or structured output) composition strategy.

From now on, ChatGPT schema also supports the optional properties and dynamic key typed object unless the strict mode configured.

Advised by @AcrylicShrimp

What's Changed

• Enhance function calling test functions by @samchon in #111
• Make ChatGPT strict mode configurable. by @samchon in #112

Full Changelog: v2.1.2...v2.2.0

v2.1.2

What's Changed

• Remove ILlmFunction.strict property. by @samchon in #109
• Also erase IHttpLlmFunction.strict property. by @samchon in #110

Full Changelog: v2.0.4...v2.1.2

v2.0.4

What's Changed

• Fix parameters separator for empty object case. by @samchon in #108

Full Changelog: v2.0.3...v2.0.4

v2.0.3

What's Changed

• Fix additionalProperties' separation problem by @samchon in #107

Full Changelog: v2.0.2...v2.0.3

v2.0.2

What's Changed

• OpenApi.IOperation["x-samchon-human"] blocking LLM function composition by @samchon in #106

Full Changelog: v2.0.1...v2.0.2

v2.0.1

What's Changed

• Add keywords (tags) by @samchon in #101
• Closes samchon/typia#1366: add discord server link by @samchon in #102
• Fix #104: conversion OpenAPI version when allOf.length is 1. by @samchon in #105

Full Changelog: v2.0.0...v2.0.1

v2.0.0

Middle layer ILlmSchema has been divided to actual LLM provider models:

• IChatGptSchema: OpenAI ChatGPT
• IClaudeSchema: Anthropic Claude
• IGeminiSchema: Google Gemini
• ILlamaSchema: Meta Llama

What's Changed

• Wiriting Gemini schema and its type checker by @samchon in #58
• Add ChatGPT schema by @samchon in #59
• GeminiConverter.schema() function by @samchon in #60
• Recursive level on LLM schemas by @samchon in #61
• LLM schemas of OpenAPI v3 and v3.1 by @samchon in #62
• Fix bug of recursive ChatGPT schema conversion by @samchon in #63
• IChatGptSchema.IOneOf.discriminator by @samchon in #64
• HttpLlm.application() bans the multipart/form-data type. by @samchon in #65
• VSCode Debugger Configuration by @samchon in #70
• Adapt typia v7 and nestia v4 for testing. by @samchon in #71
• The exact IChatGptSchema and LLM function calling applilcation schemas. by @samchon in #72
• @format uuid like description for IChatGptSchema and documetation. by @samchon in #73
• ILlmApplication.IChatGptOptions by @samchon in #74
• More example cases for the ChatGPT function calling. by @samchon in #75
• Enhance function callling prompt examples by @samchon in #76
• Enhance reference type description rule by @samchon in #78
• Completed Google Gemini schema by @samchon in #79
• Strict JSON/LLM schema of object type by @samchon in #81
• LLM schema conversion layer by @samchon in #82
• Export ILlmSchema by @samchon in #83
• Support new LLM provider, "Anthropic Claude" by @samchon in #84
• Enhance function calling test programs. by @samchon in #85
• Ban constraint fields IChatGptSchema. by @samchon in #86
• Started supporting Meta Llama schema ILlmSchema by @samchon in #87
• Update README with example codes by @samchon in #89
• Entire refactoring for concenstrate to the ILlmSchema. by @samchon in #90
• Allow additionalProperties in Claude and Llama. by @samchon in #91
• Add test cases of failures by @samchon in #92
• Fix gemini schema's oneOf detection logic. by @samchon in #93
• ILlmSchema.IParameters.additionalProperties to be false. by @samchon in #94
• Precise ILlmFunction.ISeparated<Model> by @samchon in #95
• Make constraint option to be default true by @samchon in #96
• Strong test programs for LLM schema composers and refactorings by @samchon in #97
• Try<T, E> pattern to LLM related functions by @samchon in #98
• Fix IResult.ISuccess.value. by @samchon in #99
• Completed v2 publishing by @samchon in #100

Full Changelog: v1.1.2...v2.0.0