UPtrimFrequently Asked
Questions
Everything you need to know about the early bird offer, licensing, compatibility, and getting started.
Pricing & Licensing
How long does the early bird pricing last?
The $40/yr early bird window is available during the v1.0.0 launch period. Once it closes, the same features move to $100/yr (Standard). Your license is valid for 12 months from activation.
How does annual billing work?
You pay once per year. Your license is valid for 12 months from activation. When it's time to renew, you'll get a reminder in the dashboard. If you don't renew, the proxy continues working with the free developer key features.
Do I get fewer features at the early bird price?
No. Early bird ($40/yr) and Standard ($100/yr) unlock the exact same features — unlimited memories, users, multiple backends, agent mode, Ghost Agent & Swarm Intelligence, remote access. The only difference is price. The Premium tier ($200/yr) adds 1 GB per file uploads, Knowledge Graph, Persona Engine, and more.
What's the Premium tier?
Premium ($200/yr) includes everything in Standard plus: 1 GB per file uploads (vs 25 MB per file), Memory Gardener, Knowledge Graph, Persona Engine, Predictive Context, priority support, and early access to new features. Note: Ghost Agent & Swarm Intelligence is already included in Standard.
Do I get future updates?
Your license covers all v2.x updates for the duration of your annual subscription. As long as your license is active, you get every update and patch.
Compatibility & Setup
What LLM backends are supported?
Any OpenAI-compatible API endpoint: llama.cpp, Ollama, vLLM, and hosted providers like OpenRouter. UPtrim sits between your chat UI and backend — it doesn't care what's behind the API. Hardware acceleration is fully supported: NVIDIA CUDA, AMD ROCm, and Apple MLX all work out of the box with your LLM backend.
What chat UIs work with this?
Open WebUI and SillyTavern have first-class support with automatic identity resolution. Any OpenAI-compatible client should work for basic functionality.
What platforms does it run on?
Windows, Linux, and macOS. Available as a standalone binary (PyInstaller-bundled) or run from source with Python 3.12/3.13. Works with NVIDIA CUDA, AMD ROCm, and Apple MLX hardware acceleration through your LLM backend.
How long does setup take?
Most operators are up and running in under 5 minutes. Download, point config.json at your LLM backend, connect your chat UI to the proxy endpoint, done.
Teams & Deployment
Can I use this for a team or shared deployment?
Yes. Multi-user isolation is a core feature. Each user gets their own memory space with strict identity boundaries. No memory bleed between users.
How does identity resolution work?
UPtrim reads identity headers from your chat UI (Open WebUI sends these automatically). It supports strict, required, and quarantine modes with optional HMAC signing for high-security deployments.
Can teams get custom rollout terms?
Yes. Use the contact page for enterprise rollout, volume licensing, or custom support requirements.
Support
What if I need help setting up?
The Help Center has setup guides, architecture diagrams, and configuration references. You can also reach us through the Contact page for direct support.
Is there a free version?
Yes. v1.0.0 ships with a free developer key that includes memory (30 per user), context trimming, multi-user (up to 5), unlimited file uploads (15 MB per file), and the full dashboard. The paid license removes limits and unlocks advanced features like agent mode, web search, custom prompts, and persona engine.