RiftIcon — Beta Tester Application

// Hardware Setup

1. What is your primary GPU?

2. How much VRAM does your primary GPU have?

3. How many additional devices do you have on the same WiFi network? e.g., second PC, laptop, MacBook, Android phone

0 — just my main machine 1 additional device 2 additional devices 3+ additional devices

4. List each additional device with approximate specs. Example: "MacBook Pro M3 Max, 36GB RAM" or "Old gaming PC, GTX 1070, 16GB RAM"

5. How much system RAM does your primary PC have?

6. What is your primary operating system?

// Technical Experience

7. How would you describe your comfort level with local AI tools?

Expert — I compile llama.cpp from source and fine-tune models Advanced — I regularly use Ollama, LM Studio, or KoboldCpp Intermediate — I've run a local model before but mostly use cloud Beginner — I've heard of local AI but haven't tried it Brand new — what's a GGUF?

8. Which tools have you used? Check all that apply

9. Have you ever tried to run a model that was too large for your GPU?

Yes — and I gave up or used CPU offloading (painfully slow) Yes — and I split it across multiple GPUs in one machine No — I stick to models that fit my VRAM No — I only use cloud APIs

// Use Case

10. Which description fits you best?

The Tinker — I love experimenting with AI models and pushing hardware limits for fun The Builder — I'm developing a product that needs local AI and I'm trying to avoid cloud API costs The Privacy Pro — I work under NDA, HIPAA, or in a regulated industry where data can't leave my network The Curious — I'm interested but don't have a specific use case yet

11. What would you use RiftIcon for? Check all that apply

Running large language models (70B+) locally Replacing cloud API spend (OpenAI, Anthropic, etc.) Building AI agents or multi-agent workflows on local hardware Private / air-gapped inference for sensitive data AI-assisted coding with local models Creative writing / roleplay / character AI Research and experimentation

12. How much do you currently spend per month on cloud AI APIs?

$0 — I only use free tiers or local models $1–$20 $20–$100 $100–$500 $500+

// Tester Reliability

13. How many hours per week could you dedicate to testing?

10+ hours 5–10 hours 2–5 hours Less than 2 hours

14. How would you rate your ability to write clear bug reports?

Excellent — I include steps to reproduce, logs, and screenshots Good — I can describe what happened vs. what I expected Fair — I'll say it's broken but might need guidance I've never written a bug report

15. Are you comfortable using Discord for feedback?

Yes — I'm active on Discord daily Yes — I check it a few times a week I can set it up if needed No — I prefer other channels

16. Would you share anonymized performance logs to help improve the product? tok/s, mesh topology, device specs — no personal data

Yes, absolutely Yes, but I'd want to review what's shared first No

// Final Questions

17. What is the single biggest pain point you have with running AI locally today?

18. Why would you be a great tester?

19. How did you hear about RiftIcon?

Your name

Your best email

Discord handle (optional)

Apply to be a Beta Tester