TomCon.Net's Free AI Provider Service

What Is This?

TomCon.net offers a free public service for hobbyists to experiment with AI models through an easy-to-use web interface. Think of it as your personal AI playground where you can chat with different types of artificial intelligence, analyze images, get help with coding, or just explore what modern AI can do.

Perfect for: Students, developers, researchers, and anyone curious about AI who wants hands-on experience without the technical hassle of setting up their own models.

Available AI Models

I currently host two different AI models, each designed for different types of tasks:

COSMOS-REASON (Vision AI)

Best for: Image analysis, document scanning, visual tasks

This AI can "see" and understand images you upload. Here's what it can do:

• OCR (Text Recognition): Upload a photo of a document, receipt, or handwritten note and it will extract all the text
• Image Description: Upload any picture and ask it to describe what it sees
• Visual Q&A: Ask specific questions about images ("What color is the car?" "How many people are in this photo?")
• Text-to-Speech: Optionally narrate its responses out loud

Example: Take a photo of your handwritten notes with your phone, upload it, and ask "Can you type out these notes for me?"

DEEPHERMES-8b (Reasoning AI)

Best for: Problem-solving, coding help, complex reasoning

This AI specializes in thinking through complex problems step-by-step. It's particularly good at:

• Programming Help: Debug code, explain concepts, write functions
• Math and Logic: Solve multi-step problems, explain reasoning
• Analysis: Break down complex topics into understandable parts
• Planning: Help organize projects or solve multi-part challenges

Example: "I'm getting a Python error when trying to read a CSV file. Here's my code... what's wrong and how do I fix it?"

How to Get Started

Step 1: Request Access

Contact me through one of the methods on my [contacts page]. I'll send you a personal API key within 24 hours (usually much faster).

Step 2: Log In

Visit https://ai.tomcon.net and enter your API key to access the interface.

Step 3: Choose Your Model

Select either COSMOS-REASON for image tasks or DEEPHERMES for text/reasoning tasks. (I need to manually switch between models, so let me know if you need a specific one.)

Step 4: Start Experimenting

• For images: Click the green 'add image' button to upload photos
• For text: Just start typing your questions or requests
• Responses typically take 10-30 seconds depending on complexity

Service Details

Performance & Availability

• Uptime: Best effort basis (this runs on my home server)
• Response Time: 10-30 seconds for most requests
• Concurrent Users: Limited by hardware, may queue during peak times
• Maintenance: Service may go offline for upgrades with little notice

Usage Limits

• Rate Limiting: Please be reasonable with requests (no automated spamming)
• Daily Usage: No hard limits, but excessive use may result in temporary restrictions
• Image Uploads: Standard photo sizes supported (under 10MB recommended)

Technical Specifications

• Hardware: Recycled desktop PC with Nvidia 1070ti 8GB
• Power Usage: ~200W under load, ~60W idle
• Security: HTTPS with SSL/TLS encryption for all communications
• Privacy: Quiet mode enabled (I cannot see your conversations)

Frequently Asked Questions

Q: Is this really free?

A: Yes, completely free. This is my way of giving back to the hobbyist community.

Q: What if the service is down?

A: Since this runs on my personal hardware, outages happen. Check back later or contact me if it's been down for more than a day.

Q: Can I use this for commercial projects?

A: This is intended for educational and hobby use. For commercial applications, please contact me to discuss.

Q: How do I switch between models?

A: Currently, I need to manually change models on the server. Contact me if you need a specific model or want to reserve time for a project.

Q: What are the limitations?

A: These models can't browse the internet, don't have real-time information, and shouldn't be used for medical/legal advice. Image model works best with clear, well-lit photos.

Q: My API key isn't working. What now?

A: Contact me with the email you originally used. I can regenerate your key or troubleshoot access issues.

Important Limitations

What These AIs CAN'T Do

• Browse the internet or access real-time information
• Remember conversations between sessions
• Provide medical, legal, or financial advice
• Process very large files or extremely complex images
• Guarantee 100% accuracy (always verify important information)

Best Practices

• Be specific with your questions
• For images, ensure good lighting and clear text
• Break complex requests into smaller parts
• Verify any important information independently

Privacy & Security

What I Don't Track

• Your conversation content (quiet mode enabled)
• Personal information beyond basic connection logs
• Usage analytics or telemetry data
• IP addresses beyond standard server logging

What's Protected

• All communications encrypted with HTTPS/SSL
• No conversation logs stored on the server
• No third-party analytics or tracking
• Your API key is your only identifier

Rules & Terms

Acceptable Use

• Educational and hobby projects: Encouraged
• Learning and experimentation: Perfect
• Code debugging and help: Great use case
• Image analysis for personal projects: Ideal

Not Allowed

• Uploading illegal images or content
• Asking for guidance on illegal activities
• Automated spamming or excessive requests
• Commercial use without permission
• Attempting to reverse engineer the service

My Rights

I, Thomas Conway, reserve the right to:
- Suspend or terminate access for any user
- Take the service offline for maintenance or upgrades
- Modify these terms or service features
- Discontinue the service at any time

This is a personal project I maintain in my spare time. While I try to keep it stable and available, I make no guarantees about uptime or continued service.

Get Support

Having issues or questions? Here's how to reach me:

• Check the FAQ above first
• Visit my contacts page for current contact methods
• Include "TomCon AI Service" in your subject line
• Describe the issue and include your API key if relevant

Response Time: Usually within 24 hours, often much faster.