Choosing an AI provider and model for Iris
Iris’s features inside rConfig (config analysis, drift narration, compliance narration, the optimisation assistant) run on one AI provider that you choose. It’s a decision about where your configs go, how good the answers are, and who pays. This page compares the options; the provider setup page has the steps.
rConfig docsHow to set up an AI provider in rConfig V8Connect Anthropic, OpenAI, Gemini, Azure OpenAI, Mistral, xAI, DeepSeek, Groq, Cohere, OpenRouter, Ollama or an OpenAI-compatible server to the AI Center.Read the docs
The three questions
- Where may configs go? Even with secrets obfuscated, a config describes your network in detail. Some organisations can send it to a cloud API under a data agreement; some need it to stay in their cloud tenant; some need it to stay on site.
- How long are your configs? Config analysis sends the whole config on the first chat message, up to about 100,000 characters. Long firewall and core router configs need a capable model with a large context window.
- Who pays, and how? Cloud providers bill per token. A self-hosted model costs hardware and power, and nothing per question.
Options compared
| Option | Providers | Configs go to | Answer quality on long configs | Cost model | Design guide |
|---|---|---|---|---|---|
| Public API | Anthropic, OpenAI, Google Gemini, Mistral, xAI (Grok), DeepSeek, Groq, Cohere, OpenRouter | The provider, under its data terms | Highest | Per token | Public API |
| Private cloud | Azure OpenAI (your deployment) | Your Azure tenant and region | High, depends on the model you deploy | Per token, on your Azure bill | Azure OpenAI |
| Self-hosted | Ollama (local), OpenAI-Compatible (vLLM, LM Studio, LiteLLM) | Nowhere outside your network | Depends on model size and hardware | Hardware | Self-hosted |
| Split | A self-hosted model for Iris, cloud agents over MCP for engineers who need more | Iris: on site. Agents: the agent's provider | Mixed | Both | Split |
Providers offered in rConfig 8.4.2.
rConfig’s Privacy tab classifies whatever you pick. Ollama or OpenAI-compatible on localhost, a private address, or a .local or .internal host counts as On-Premises. Named cloud providers count as Cloud Provider. An OpenAI-compatible URL anywhere else counts as External API Endpoint. See where your data goes.
Providers and models in rConfig
Every provider rConfig offers, and the models AI Settings lists for each. For providers with a list, you can also type any model ID the provider supports. This table is generated from rConfig’s own model catalogue, so it changes when rConfig’s list does.
| Provider | Runs | Models in AI Settings | Used if no model is set |
|---|---|---|---|
| Anthropic | Vendor cloud | claude-sonnet-5-5 (recommended), claude-opus-5-5, claude-haiku-5-5, claude-opus-4-8, claude-sonnet-4-6, claude-haiku-4-5-20251001, or enter any model ID | claude-haiku-4-5-20251001 |
| OpenAI | Vendor cloud | gpt-6.1-sol (recommended), gpt-6-astra, gpt-6-luna, gpt-4o, gpt-4o-mini, o3-mini, o4-mini, or enter any model ID | gpt-4o-mini |
| Google Gemini | Vendor cloud | gemini-3.8-flash (recommended), gemini-3.1-flash-lite, gemini-2.5-pro, gemini-2.5-flash, or enter any model ID | gemini-3.1-flash-lite |
| Azure OpenAI | Vendor cloud | You enter the deployment name from your Azure deployment | gpt-4o |
| Ollama (local) | Your network | You enter the model name. Suggested: qwen3.5:4b, qwen3.5:0.8b, llama3.1:8b | llama3 |
| OpenAI-Compatible | Your network | You enter the model name | gpt-4o-mini |
| Mistral | Vendor cloud | mistral-large-2512 (recommended), mistral-medium-3-5, mistral-small-2603, or enter any model ID | mistral-small-2603 |
| xAI (Grok) | Vendor cloud | grok-4.7 (recommended), grok-4.20-non-reasoning, or enter any model ID | grok-4.20-non-reasoning |
| DeepSeek | Vendor cloud | deepseek-flash (recommended), deepseek-v4-pro, or enter any model ID | deepseek-flash |
| Groq | Vendor cloud | openai/gpt-oss-120b (recommended), openai/gpt-oss-20b, or enter any model ID | openai/gpt-oss-20b |
| Cohere | Vendor cloud | command-a-03-2025 (recommended), command-a-plus-05-2026, command-r7b-12-2024, or enter any model ID | command-r7b-12-2024 |
| OpenRouter | Vendor cloud | anthropic/claude-sonnet-5.5 (recommended), anthropic/claude-fable-5.1, anthropic/claude-haiku-4.5, or enter any model ID | anthropic/claude-haiku-4.5 |
From rConfig 8.4.2's model catalogue, synced 2026-11-12. The list follows rConfig automatically; providers retire models, so check the dropdown in AI Settings for your version.
Choosing a model
- For config analysis, use the strongest model you can. Reading a 3,000-line firewall config and spotting the one rule that shadows another is hard. Small models get the easy questions right and the hard ones confidently wrong.
- Context window matters as much as model size. A first chat message can carry the config, rConfig’s background knowledge (up to 14,000 characters) and your question. If the model’s context window is too small, the request fails or is silently cut. For self-hosted models see Sizing a local model.
- Answer length is capped twice. Each feature has its own ceiling (4,096 output tokens for config analysis, drift narration and the optimisation assistant, 2,048 for compliance narration and rule generation). The Max Output Tokens setting can lower that, never raise it.
- Lower temperature suits config work. Focused, repeatable answers are what you want. Leave temperature at the default or lower it.
Switching later
You can change the active provider at any time in AI Center → AI Settings. Many teams start on a public API to see what Iris can do, then move to Azure OpenAI or a self-hosted model once they know which features they use. The features and prompts stay the same; only answer quality and cost move.
Related reference
- Set up an AI provider
- AI Center governance, audit trail and privacy
- How rConfig grounds AI responses
Frequently asked questions
Which AI providers does rConfig support?
Anthropic, OpenAI, Google Gemini, Azure OpenAI, Ollama, and any server that speaks the OpenAI chat API (vLLM, LM Studio, LiteLLM). From V8.4.2 Mistral, xAI, DeepSeek, Groq, Cohere and OpenRouter are also offered as providers.
Can rConfig's AI features run without sending configs to the cloud?
Yes. With Ollama or a self-hosted OpenAI-compatible server on your network, rConfig's Privacy tab classifies the provider as on-premises and configs never leave your environment.
Does the provider I pick in rConfig affect Claude Code or Cursor?
No. The provider setting only drives the Iris features inside rConfig. Agents connected over MCP use their own model.