Multi-Provider AI Engine & API Key Management
Translucent features a universal streaming AI dispatcher that seamlessly connects to Google Gemini, Groq, OpenAI, DeepSeek, OpenRouter, and local Ollama instances. You can configure and switch models on the fly with per-provider key retention.
Supported AI Providers & Key Endpoints
Direct links to developer portals and recommended model identifiers for each engine.
| AI Provider | Default & Recommended Models | Direct Key Link | Speed & Latency | Quick Action |
|---|---|---|---|---|
Google GeminiOfficial / Default | gemini-3.8-flashgemini-2.5-flashgemini-3.5-flash-lite | aistudio.google.com/apikey | ~250ms (Ultra fast) | |
Groq (Free & Ultra Fast)Fastest In The World | llama-3.3-70b-versatilellama-3.1-8b-instantmixtral-8x7b-32768 | console.groq.com/keys | ~80ms (Real-time LPUs) | |
OpenAI (GPT-4o)Industry Benchmark | gpt-4ogpt-4o-minio3-mini | platform.openai.com/api-keys | ~400ms | |
DeepSeekReasoning Powerhouse | deepseek-chatdeepseek-reasoner | platform.deepseek.com | ~350ms | |
OpenRouterAll-In-One Unified Gateway | meta-llama/llama-3.3-70b-instructdeepseek/deepseek-chatanthropic/claude-3.5-sonnet | openrouter.ai/keys | ~300ms | |
Custom / Local (Ollama)100% Offline & Private | llama3mistraldeepseek-r1qwen2.5-coder | http://localhost:11434 | Local GPU dependent (~20-100ms) |
Select AI Engine
Google Gemini
Official / DefaultGoogle's flagship multimodal model designed for complex reasoning, instant code generation, and multi-image vision analysis. Includes built-in auto-failover in Translucent during peak demand.
Setup Steps in Translucent
How Translucent Handles Gemini 503 High Demand Spikes
When Google's public endpoints return 503 ServiceUnavailable ("model experiencing high demand") or 429 RateLimit on gemini-3.8-flash, Translucent automatically re-routes the prompt to gemini-2.5-flash in milliseconds.
QUICK REFERENCE
Essential Stealth Hotkeys
Global Vanish / Toggle
Instantly shows or hides the entire Translucent desktop overlay from any active window or game.
Stealth Screen Snip
Select any region on screen to send directly to Gemini or GPT-4o vision for instant analysis.
Ghost Mode (Click-Through)
Enables mouse clicks and typing to pass straight through the window into background applications.
Mini Floating Bubble
Minimizes the assistant into an ultra-compact 54x54 floating glowing bubble pinned on top.
Dual Split View
Locks the stealth Chromium browser tab alongside the AI assistant side-by-side.
AutoTyper Assistance
Automatically types answers or selected code directly into your active code editor.
Need Setup Help or Have Questions?
Our engineering team provides direct one-on-one customer support on Telegram, YouTube, and via email (support.translucent@gmail.com). Whether you need help setting up custom API keys, configuring VMware, or instant license assistance, message us anytime.
