Skip to main content
Noxus provides a unified interface to access leading AI models from multiple providers. Connect your provider credentials once and manage all your models, embeddings, and presets from a single settings page.

Model Configuration

All model and provider management lives under Settings > Models. The page is organized into these tabs:
Configure and manage your AI provider connections. Each provider shows its name, how many LLMs and embeddings are linked, health status, and an active toggle.
Providers tab showing connected providers with health status

Decision models

A decision model reads a text and answers fixed questions about it (yes or no, one of several options, or a level on a scale) with a probability for every answer. It never writes text, so it is fast and cheap. To enable one, open the Decision models tab, click Add decision model, pick the model and a provider you have already connected. Noxus asks it one real question before enabling the link, the same way LLMs are checked. Each call is billed as a decision model request in usage reports. The Jev reranker is listed on the Rerankers tab and links to TypeSafe or OpenRouter. A knowledge base uses it when you pick it as its reranker. If it fails, the next reranker in the chain runs, and a search never fails because of reranking.

Voice models

Voice models turn speech into text or text into speech. They are a catalogue of their own, so they never appear in a language model picker or a preset. To enable one, open the Voice models tab, click Add model in the table for its type and pick a provider you have already connected. Noxus makes one real call before enabling the link (a one-word synthesis or a one-second transcription), and shows the provider’s own reason when it fails. Click a model to see its voices, the formats it writes or the audio it accepts. Usage is billed per minute, per character or per audio token, the way the vendor bills it.
  • AWS: Amazon Transcribe and Polly need a Bedrock provider that signs in with AWS credentials. A Bedrock API key only reaches Bedrock models.
  • Chirp 3: the Google Cloud project behind the Vertex AI provider must have the Cloud Speech-to-Text API enabled.
  • ElevenLabs: the voice dropdown lists your ElevenLabs account’s own voices as well as the premade ones.

Supported Providers

Noxus supports a wide range of cloud AI providers. Click Add provider to see the full list and connect a new one.
Add provider dialog showing all available providers

Azure OpenAI behind an API Management gateway

The Azure OpenAI Endpoint is the OpenAI v1 address of your resource, for example https://my-resource.openai.azure.com/openai/v1/. If your organization routes Azure traffic through an Azure API Management gateway, enter the gateway’s v1 address instead and put the gateway’s subscription key in API key. Gateways built from Microsoft’s AI Citadel accelerator serve it at /models/, for example https://my-gateway.azure-api.net/models/. Chat requests go to that address. Embeddings go to the gateway’s /openai/deployments/<deployment>/embeddings route, the only route Azure serves embeddings on, so the gateway must expose it as well.

Vertex AI regions and data residency

Pick Multi-Region (EU) or Multi-Region (US) to keep model requests inside that area. They work with every auth mode, including API key. Only models Google serves in a multi-region answer there, such as Gemini 3.5 Flash and Gemini 3.5 Flash-Lite. Global can route requests outside the EU and US, so add it only when your data-residency rules allow it. When you add a model that none of the provider’s regions serves, a dialog lists the model’s regions and where each one sends requests. Pick one and Noxus adds it to the provider and checks the model again.
Dialog offering the EU multi-region, Global and the US multi-region for a model the provider's regions do not serve
Noxus continuously adds support for new providers. If you don’t see a specific provider listed, it may still be supported via OpenRouter or our plugin system.

Connecting a Provider

1

Open Model Settings

Navigate to Settings > Models. You’ll land on the Providers tab.
2

Add a provider

Click Add provider in the top right. Select Model or Observability from the left sidebar, then choose your provider.
3

Enter credentials

Fill in your API key or complete the authentication flow. For providers like Vertex AI and Bedrock, choose your auth mode and configure regions.
4

Test the connection

Noxus automatically tests your credentials. For multi-region providers (Vertex, Bedrock), each region is tested individually with a step-by-step health pipeline.
5

Enable models

Switch to the LLMs or Embeddings tab and toggle on the models you want to make available.

Provider Details

Click any provider row to open its details drawer. Here you can see:
  • Connection info — name, active toggle, linked model counts
  • Health status — last check time, current status, and a Test connection button
Provider details drawer showing health status and test connection

AWS Bedrock regions

Bedrock serves Claude through cross-region inference profiles, and each region can call only some of them. For every region you select, Noxus calls the profile that region lists for the model: A region with no profile for a model is skipped, and the next provider in the preset answers instead. Some models have no profile for your geography: Claude Fable has none in the EU, and Seoul, Singapore and São Paulo have none for current Claude models. Turn on Global inference on the provider to use the model’s global. profile in those regions. AWS may then process the request in any of its commercial regions, so leave it off when requests must stay in your geography.

Health Monitoring & Model Lifecycle

Noxus continuously monitors your provider connections and model availability through automated health checks that form the model lifecycle system.

Provider Health Statuses

These statuses appear on the Providers tab and reflect the overall health of a provider connection:

Model Health Statuses

Individual model–provider links (shown on the LLMs and Embeddings tabs) have their own lifecycle:

Connection Testing

When you test a provider connection, Noxus runs a multi-step health pipeline that validates each aspect of your configuration:
  • Authentication — are your credentials valid?
  • Permissions — do you have the right access level? (Vertex IAM, Bedrock invoke permissions)
  • Regional access — for multi-region providers, each configured region is tested individually
Each step reports pass, fail, or skip, with actionable hints when something goes wrong. Results stream in real-time so you can see progress as each step completes.

Automatic Lifecycle Management

Noxus runs periodic health checks on all provider connections and model links in the background. Based on results, models transition automatically between lifecycle states:
  1. A healthy model starts failing → after several consecutive failures it becomes suspended and is excluded from routing.
  2. If failures persist → the model is deactivated and fully removed from routing.
  3. Periodic recovery probes test deactivated models → if a probe succeeds, the model is restored to healthy.
This means transient provider outages are handled automatically without manual intervention. You can also manually reactivate a deactivated model at any time from the LLMs or Embeddings tab.

Vendor Retirements

Vendors announce the date they stop serving a model ahead of time. Noxus records that date in the catalogue. In the LLMs tab the model’s Status badge reads “Retires 20 Oct” and the Status filter has a Retiring entry; a banner above the table lists everything retiring in the next 30 days. In the model picker and the model explorer a calendar icon next to the model’s name shows the same date on hover. Move flows and agents that use the model to a replacement before that day: once it passes, the model is marked deprecated and dropped from the default presets.

Model Selection in Flows and Agents

When configuring an AI node in a flow or an agent, you select models through the Model tab in the configuration drawer.

Preset Selection

By default, nodes use a model preset — a named bundle of models tried in priority order. If the first model encounters an error, the next one is used automatically.
Model picker showing a preset with fallback chain
Click the preset dropdown to switch between presets or choose Custom models for manual selection.
Preset dropdown showing available presets

Custom Model Selection

When you choose Custom models, a model browser opens showing all available models across your connected providers. Each model appears once per provider connection, so if you’ve connected both OpenAI and Azure OpenAI, you’ll see separate entries for each.
Model selection modal with search, filters, speed and quality gauges
The model browser includes:
  • Search — find models by name
  • Filters — narrow by provider, speed, quality, location, and capabilities (vision, function calling, reasoning, etc.)
  • Speed & Quality gauges — visual indicators to compare models at a glance
  • Model details — hover over a model to see its full specs: speed (tokens/sec), quality score, context window, release date, and capabilities
  • Fallback chain — selected models are ordered by priority. The first model is the default; others are fallbacks

Local & Custom Models

For organizations with strict data residency requirements or proprietary models, Noxus offers deep integration for self-hosted infrastructure.

Seamless Local Integration

  • Private Endpoints: Connect to on-premises inference servers (e.g., vLLM, Ollama, TGI) via secure private networking using custom base URLs on OpenAI-compatible providers.
  • Unified Interface: Local models appear alongside cloud providers, allowing for seamless switching in flows and agents.

Custom Model Providers

You can extend the platform to support any proprietary or specialized model through our plugin system.
  • Custom Inference: Build plugins for internal model servers or niche providers.
  • Fine-tuned Models: Easily integrate your organization’s fine-tuned models into the standard workflow.

Observability Providers

In addition to model providers, you can connect observability backends to trace and monitor all AI model calls. See Observability Providers for details.

Model Presets

Configure named model bundles for consistent selection across flows.

Model Selection Guide

Choose the right model for your task based on quality, speed, and cost.

Observability Providers

Connect tracing backends for full visibility into model calls.

Security Architecture

Understand how your API keys and model data are protected.