Core Experience
PlexonProvider
One subscription, every capability. Code, generate images, create videos, and analyze documents, all through a single provider that handles everything behind the scenes.
Just start building.
Plexon is the default provider for every new installation. No API keys to find, no models to choose, no rate limits to worry about. Your subscription covers everything: coding, image generation, video creation, and document analysis.
One prompt, multiple capabilities
When your request spans task types, Plexon automatically decomposes it into specialized sub-agents. Each runs with the right model, and results are synthesized into a single response.
Code Agent
The default agent for all text tasks. Writes code, debugs, refactors, plans architecture, analyzes documents, and answers questions.
Image Agent
Generates images from text descriptions. Automatically invoked when your prompt includes visual creation: hero images, diagrams, illustrations, and more.
Video Agent
Creates videos from text or image descriptions. Handles product demos, animations, and visual storytelling. Supports image-to-video for reference-based generation.
"Build a landing page for a dog walking app and generate a hero image for it"
Plexon decomposes this into two parallel sub-agents: a Code Agentbuilds the HTML/CSS, and an Image Agent generates the hero image. Results are synthesized into a single, coherent response.
Auto Routing
By default, Plexon sends every request to the coding model: zero overhead, instant response. But if you frequently switch between coding, image, and video tasks in the same conversation, toggle on Auto Routing.
When enabled, an AI classifier analyzes each message and routes it to the best model automatically. It adds a small amount of latency but ensures every request uses the optimal model for the task.
Built-in resilience
Behind every Plexon request is a pool of API keys with automatic failover, per-user rate limiting, and real-time health monitoring.
Automatic Key Rotation
When a key hits its rate limit, the system immediately rotates to another available key. Exponential backoff keeps rate-limited keys in cooldown while fresh ones take over.
Per-User Rate Limits
Fair usage enforcement prevents any single user from exhausting the pool. 4,500 text requests per 5 hours, 50 images per day, and 10 videos per day.
Usage Tracking
Check your remaining usage anytime. The usage endpoint shows requests used, limits, and exact reset times for every category.
BYOK Compatible
Prefer your own API keys? Switch to any provider (Anthropic, OpenAI, Google, or others) and enter your key. Your keys always take priority over the server pool.
Plexon Provider is included with every subscription.
View PlansMore Capabilities
1M Context
Full project understanding with a 1M token context window. Feed entire codebases, long documents, and complex specifications.
Prompt Caching
The Plexon identity and system prompts live in the static cached block: up to 90% cost reduction across conversation turns.
Branded Experience
The AI identifies as Plexon. Error messages, system prompts, and responses never expose the underlying infrastructure.
Subscription Tiers
Pro subscribers get standard-tier keys. Premium subscribers get priority-tier keys with reserved capacity for faster response times.