Blog

All Blog Posts  |  Next Post  |  Previous Post



Introducing TMS AI Studio: Your Complete AI Development Toolkit for Delphi

Thursday, July 24, 2025

We're thrilled to announce the release of TMS AI Studio, a revolutionary new product that brings the power of AI directly into your Delphi development workflow. Building on TMS Software's tradition of innovative developer tools, TMS AI Studio represents our most ambitious step yet into the future of AI-powered application development.

TMS Software Delphi  Components

What is TMS AI Studio?

TMS AI Studio is a comprehensive AI development toolkit that lets you effortlessly add AI capabilities in your Delphi apps. It's designed to help Delphi developers seamlessly integrate AI functionality into their applications, from simple chatbots to complex AI-powered business solutions.

At its core, TMS AI Studio implements the Model Context Protocol (MCP), a standardized framework for building AI tool integrations. This means your applications can communicate with AI services using industry-standard protocols, ensuring compatibility and future-proofing your investments.

TMS Software Delphi  Components

With TMS AI Studio, you can:

  • Add chatbots to your applications
  • Perform intelligent information analysis and generation
  • Offer natural language interfaces for user interaction
  • Create agentic flows with custom MCP servers and clients
  • Process images, PDFs, and audio files to extract information
  • Perform OCR, text-to-speech, and speech-to-text tasks
  • Define custom tools and enable LLM-based function calling
  • Seamlessly switch between multiple LLM providers
In short, with TMS AI Studio, you only feel yourself limited by your own imagination.

Why TMS AI Studio

Software is undergoing a revolution. Not only do developers expect to harness AI to create applications faster, but users also demand smarter, more intuitive software experiences. Don’t get left behind—make sure your applications deliver a state-of-the-art, AI-powered experience.

TMS Software Delphi  Components

Key Features and Components

TTMSMCPCloudAI component to abstract LLM usage

For developers who need immediate AI capabilities the TTMSMCPCloudAI component offers a streamlined approach. This component provides direct access to multiple LLM services with context management, context attachments, function calling, web search and customizable settings. The basic usage starts with simple code like:

TMSMCPCloudAI.Context.Text := 'What do you know about Delphi?'; TMSMCPCloudAI.Service := aiOpenAI; TMSMCPCloudAI.Settings.Temperature := 0.7; TMSMCPCloudAI.Execute; TMSMCPCloudAI.OnExecute := MyResponseHandler;

The component handles API key management, supports custom tools with parameter validation, and provides detailed response information including token usage and model details. It offers an interface to add different kind of files to analyse or augment the generated responses. Whether you're building a simple chatbot or complex AI-powered business logic, TTMSMCPCloudAI makes it straightforward to integrate AI capabilities directly into your applications.

TMS Software Delphi  Components

Supported AI providers include OpenAI (GPT models), Google Gemini, Anthropic Claude, Grok (xAI), Perplexity, Mistral AI, DeepSeek, and Ollama for local inference.

Complete MCP Implementation 

TMS AI Studio provides a full Delphi implementation of the Model Context Protocol, supporting both Server and Client over STDIO and HTTP/SSE transport layers. 

Powerful Server Components

The TTMSMCPServer component handles all MCP protocol communication, while TTMSMCPTool lets you define custom tools that AI clients can invoke. Need to share data? The TTMSMCPResource component registers and manages resources accessible via the server:

// Create MCP Server
    Server := TTMSMCPServer.Create(nil);
    Server.ServerName := 'WeatherServer';
    Server.ServerVersion := '1.0.0';
Tool := TTMSMCPTool.CreateBuilder
         .Name('get_weather')
         .Description('Get current weather information for a specified city')
         .ExecuteCallback(HandleWeatherTool)
         .ReturnType(ptJSON)
         .AddProperty
         .Name('city')
         .Description('The city name to get weather for')
         .PropertyType(ptString)
         .Required(False)
         .&End
       .Build;

Server.Tools.Add(Tool);

Server.Start;
Server.Run;

Create reusable prompt templates with TTMSMCPPrompt for consistent AI interactions across your application.

MCP Client component to build your own Delphi powered MCP clients

The TTMSMCPClient connects to multiple AI services seamlessly and handles the communication with the MCP servers. This includes Delphi created MCP servers as well as 3rd party MCP servers creating with other development languages.
Here is a code snippet showing how to create an MCP client in code and connecting it to one MCP server:

var
  client: TTMSMCPClient;
  server: TTMSMCPCLientServerItem;
begin
  Client := TTMSMCPClient.Create(Self);
  Client.LLM.APIKeys.OpenAI := 'your-api-key';
  Client.LLM.Service := aiOpenAI;
  server := TTMSMCPClient1.Servers.Add;
  server.Command := 'path-to/my-mcp-server.exe';
  //Optional arguments
  server.Args.Add('-myarg1');
  server.Args.Add('-myarg2');
  //Optional environment variables
  server.EnvironmentVariables.Add('MYVAR=myvalue');
end ;

Developer-Friendly Tools

Function calling support comes with built-in parameter validation, automatic connection management handles MCP servers behind the scenes, and comprehensive logging helps with debugging. The included STDIO Bridge connects standard MCP clients to your Delphi visual VCL or FMX applications effortlessly.

Rich Learning Resources

To help you get the most out of TMS AI Studio, we've prepared an extensive collection of resources:

Blogs

Start reading our blogposts and get a head start using the TMS AI Studio. 

Whitepaper

Download our detailed white paper: "Add AI Superpower to Your Delphi & C++Builder Apps".

This whitepaper shows how to bring powerful AI features into your Delphi and C++Builder applications. Learn how to connect to popular AI models like ChatGPT, Claude, and Gemini to make your software smarter and more intuitive. Explore how to process images and documents, build intelligent assistants, run natural language queries, and connect AI to your own app logic. With clear examples and practical steps using TMS Software components, you will see how simple it is to create modern, AI-powered desktop applications that stand out.

Video Tutorials


All-Access Users - You're Already Covered!

Great news for All-Access subscribers! TMS AI Studio is included at no additional cost as part of your existing All-Access subscription. This means you can start exploring AI development immediately without any additional investment.

All-Access users get:

✅ Full TMS AI Studio license
✅ Access to all components and tools

Important Installation Notice

As a tool bringing intelligence to your apps, TMS AI Studio expects an intelligent, reliable and flexible tool for installation of its components in your IDE. Therefore, TMS AI Studio is exclusively available through our new TMS Smart Setup installation system. 

Note that TMS AI Studio is available for Delphi or C++Builder 11 and newer versions. 

Start Building the Future Today

The AI revolution is here, and with TMS AI Studio, Delphi developers are perfectly positioned to be at the forefront. Whether you're building intelligent business applications, creating AI-powered tools, or exploring the possibilities of the Model Context Protocol, TMS AI Studio gives you everything you need to succeed.

Ready to transform your development process?

Starting at 225 EUR for a single developer license, TMS AI Studio offers exceptional value for the comprehensive AI development toolkit it provides.

Get TMS AI Studio Now →

Join the growing community of developers who are already building tomorrow's AI applications with TMS AI Studio. The future of intelligent software development starts here.

TMS Software Delphi  Components


TMS AI Studio requires Delphi 11.0 or higher. 



Bradley Velghe


  1. Add AI superpower to your Delphi & C++Builder apps part 1

  2. Add AI superpower to your Delphi & C++Builder apps part 2: function calling

  3. Add AI superpower to your Delphi & C++Builder apps part 3: multimodal LLM use

  4. Add AI superpower to your Delphi & C++Builder apps part 4: create MCP servers

  5. Add AI superpower to your Delphi & C++Builder apps part 5: create your MCP client

  6. Add AI superpower to your Delphi & C++Builder apps part 6: RAG

  7. Introducing TMS AI Studio: Your Complete AI Development Toolkit for Delphi

  8. Automatic invoice data extraction in Delphi apps via AI

  9. Creating an n8n Workflow to use a Logging MCP Server



This blog post has received 12 comments.


1. Thursday, July 24, 2025 at 6:12:52 PM

Great Looking forward to using this

Beryllium ehf


2. Friday, July 25, 2025 at 10:12:05 AM

Is there a setting for context window size?

Alexander Pastuhov


3. Friday, July 25, 2025 at 10:24:44 PM

Do you plan a lazarus/freepascal version?

Yoni


4. Saturday, July 26, 2025 at 11:31:13 AM

Is ANY of this available on D10.4.2 ?

wiz


5. Saturday, July 26, 2025 at 12:23:35 PM

Hi, only available starting from RAD Studio 11

Pieter Scheldeman


6. Saturday, July 26, 2025 at 12:58:05 PM

Hmmm ... why''s that? Are there some technical improvements to D11 missing from earlier versions? I stopped upgrading b/c I haven''t noticed anything that seemed significant.

wiz


7. Saturday, July 26, 2025 at 4:46:43 PM

Technical language specific reasons as well as the introduction for platform specific compiler defines.

Pieter Scheldeman


8. Thursday, July 31, 2025 at 4:52:09 AM

Can not use a custom OpenAI compatible API BaseURL? Usage is restricted to the official API endpoint address. Third-party API endpoints are not supported. Please add this feature.

silkwa


9. Saturday, August 23, 2025 at 1:43:46 AM

Hi, I''m evaluating (v1.3.0.0) TTMSFNCCloudAI.

I get the response.ServiceID and then I would like to set the previous_response_id = serviceID, so that the conversation can continue.
Using OpenAI. And I''m using the CustomOptions (Settings) to include the previous_response_id.
But I get an access violation when I try to do so after Execute.

Is the CustomOptions used to send specific parameters?

How do I use OpenAI and have the conversation continue? (i.e. keep the state).
According to OpenAI''s API documentation, we can use previous_response_id to continue conversation.

Joe

Joe


10. Tuesday, August 26, 2025 at 11:01:13 AM

Are you using the OpenAI GPT-5 model?

You set this CustomOptions with
"previous_response_id": "'' + PrevID + ''";
?

as this should normally work

Bruno Fierens


11. Tuesday, August 26, 2025 at 2:49:50 PM

Ok, the previous_response_id now works (gpt-4.1) after I removed the curly brackets (like a json object). Thanks.

But gpt-5 doesn''t work ... however, I''m using v1.3.0.0 of that component which I downloaded several weeks ago.
Would it work with the new version?

Joe


12. Tuesday, August 26, 2025 at 3:44:57 PM

We added support for GPT-5 in TMS FNC Cloud Pack v3.6 released last week.

Bruno Fierens




Add a new comment

You will receive a confirmation mail with a link to validate your comment, please use a valid email address.
All fields are required.



All Blog Posts  |  Next Post  |  Previous Post