Nyquest Platform Documentation

Nyquest is a multi-model AI workspace. One interface, models from every major provider, transparent pricing, built-in cost control.

This documentation covers the Nyquest App at app.nyquest.ai — the primary product. For questions or help, join the Discord.

The Workspace

The Nyquest workspace is where you interact with AI models. It includes a chat interface, conversation sidebar, model selector, and settings panel.

Layout

• Sidebar — conversation history, search, new chat
• Chat area — message input, model responses, streaming output
• Model selector — pick a model or use smart routing
• Inspect panel — per-request details (tokens, cost, model, latency)
• Settings — themes, BYOK keys, budgets, account

The workspace is fully responsive. Same features on desktop, tablet, and phone.

Sending Your First Message

Open app.nyquest.ai. Type a message and press Enter. By default, smart routing selects the best model for your prompt. The response streams back in real time.

Every message creates a conversation with an auto-generated title. Conversations are saved automatically and appear in the sidebar.

Smart Routing

When no model is pinned, Nyquest automatically routes your prompt to the best available model based on tiered logic.

How it works

• Tier 1 — premium reasoning models (Claude Opus 4.6, GPT-5.4, Gemini 3.1 Pro). Used for complex, multi-step, or high-stakes prompts.
• Tier 2 — fast general-purpose models (Claude Sonnet 4.6, GPT-5.4 Mini, Gemini 3 Flash). Used for standard questions and tasks.
• Tier 3 — cost-efficient models (Haiku, Llama, Mistral Small). Used for simple lookups and bulk work.

Routing considers prompt complexity, token count, and cost-awareness. You can override routing at any time by pinning a model.

Model Pinning

Pin a specific model to a conversation. Every message in that thread goes to the pinned model — no routing, no switching.

To pin a model, use the model selector dropdown above the chat input. Select any available model from the supported providers. The pinned model is displayed in the conversation header.

To unpin, select "Smart Routing" from the model selector. The conversation returns to automatic routing.

Supported Models

Nyquest supports models from all major providers, accessible via BYOK or hosted routing:

All models use standard provider pricing with no markup. Hosted requests deduct from your prepaid wallet. BYOK requests are always free through Nyquest. See Pricing for rates.

Routing Tiers

Smart routing uses three tiers based on prompt characteristics:

Conversations & Memory

Every conversation is stored with full message history. Conversations persist across sessions — close the tab, come back later, pick up where you left off.

Features

• Auto-titles — Nyquest generates a title from the first message
• Search — find any past conversation by keyword
• Multi-turn context — recent conversation history (up to 20 messages) is included in each request so the model has context
• Context management — long conversations are automatically trimmed to stay within model context windows without losing critical instructions

Web Grounding

Nyquest includes built-in web search. When enabled, live search results are injected into your prompt context before the model sees it.

How to use

• Toggle web grounding on/off in the chat interface
• When enabled, prompts about current events, recent data, or real-time information will include web search context
• No API keys needed — search runs on Nyquest infrastructure
• Results are injected transparently; the model responds with grounded information

Streaming

All model responses stream in real time via Server-Sent Events (SSE). Tokens appear as they're generated — no waiting for the full response to complete.

Streaming works across all supported models and all providers. There is no configuration needed.

Inspect Panel

The inspect panel shows per-request details for every message:

• Model used — which model handled the request
• Tokens — input and output token counts
• Cost — estimated cost for the request
• Latency — time to first token and total response time
• Routing decision — why this model was selected (if smart routing)
• Optimization — token reduction applied by the optimization engine

BYOK (Bring Your Own Keys)

BYOK lets you use your own API keys for any supported provider. Your requests go directly to the provider under your account and billing.

Setup

• Go to Settings → Providers
• Toggle BYOK on for any provider
• Paste your API key
• The key is encrypted immediately with AES-256-GCM

How it works

• With BYOK enabled, prompts are optimized by Nyquest, then sent directly to the provider using your key
• You pay the provider's published rate (or your negotiated rate)
• Nyquest adds zero token markup
• Keys are encrypted at rest, decrypted only at dispatch, never logged

Supported providers

OpenAI, Anthropic, Google, xAI. Add or remove keys at any time in Settings.

Themes & Appearance

The workspace ships with a full theme engine. All themes are free — always included, no paywall.

Available themes

Midnight (default dark), Aurora (purple), Glacier (light), Ember (warm dark), Terminal (green-on-black), Ocean (deep blue), Cream (warm light), Neon (cyberpunk). More themes added regularly.

Customization

• Theme selector — switch themes with one click in Settings
• Design tokens — each theme is built on CSS variables (background, surface, text, accent, border, sidebar)
• Layout modes — sidebar layout, full-width chat, compact mode
• Background styles — solid colors, gradients, mesh patterns, noise textures
• Dark / Light — both modes supported across all themes

Budget Controls

Nyquest includes optional daily budget controls. Set a spending limit, and Nyquest enforces it automatically.

How billing works

• Hosted models — requests deduct from your prepaid wallet balance. Fund via Stripe ($5–$500).
• BYOK models — always free through Nyquest. You pay the provider directly with your own key.
• When your wallet balance reaches zero, hosted requests pause. BYOK continues to work.
• Add funds any time. No subscriptions, no monthly commitments.

Optional daily limits

• Set a daily budget at any amount in Settings
• Usage is tracked in real time — per request, per model, per conversation
• When the budget is reached, requests pause until the next day
• Budgets reset every 24 hours
• You can adjust or remove your budget at any time
• If no budget is set, usage is unlimited

Budgets apply across all models using standard pricing. There are no fixed tiers — you decide what makes sense for your usage.

Account & Auth

Nyquest uses token-based authentication. Sessions are secure and persistent across browser restarts.

• Sign up — create an account at app.nyquest.ai
• Sessions — secure tokens with automatic refresh
• Security — industry-standard password hashing, AES-256-GCM key encryption
• Data — conversations and settings stored in persistent encrypted storage