Skip to main content
Each sureops agent can use a different LLM model. By default, all agents inherit the org-level default model. You can override this per agent and per environment — for example, using a more capable model for the Diagnosis Specialist in production while using a faster model in staging.

​
Default behavior

When you first set up sureops, all agents use the org-level default model. This model is configured at Settings → AI Agents, in the AI Defaults card (Default Provider and Default Model). The AI Agents tab requires an environment to be selected before it’s usable, so in practice this card almost always shows as Environment AI Defaults rather than the org-level version — it’s the same card, and its title just reflects whichever scope you’re currently viewing. Changes to the default immediately apply to all agents that haven’t been individually overridden. The model-picker and per-agent override flow described on this page applies to BYOK organizations specifically. Demo and trial organizations already get a fully locked model-selection experience today — there’s no model dropdown at all, and only prompts, tasks, and skills stay configurable per agent. The same lock will also apply to managed-mode organizations once that mode launches (see BYOK vs Managed LLM).

​
Overriding a specific agent’s model

1

Open agent settings

Go to Settings → AI Agents and select the agent you want to configure.
2

Toggle off inheritance

On the agent’s configuration card, the “Use Global” toggle is on by default — the agent inherits the org (or environment) default model. Turn it off to break inheritance and set a model specifically for this agent.
3

Select a model

Choose from the available models in the dropdown — the set sureops has validated for agent use with your BYOK provider, not every model your provider offers.
4

Save

Click Save. The change takes effect on the next incident that uses this agent.
To revert to the inherited default, toggle “Use Global” back on.

​
Per-environment overrides

An agent can use a different model in different environments. For example:
  • Diagnosis Specialist in production: a high-capability model for accuracy on critical incidents
  • Diagnosis Specialist in staging: a faster, lower-cost model for exploratory use
To configure per-environment:
  1. Switch to the environment you want to configure using the app-level environment context switcher (the breadcrumb near the top of the app) — the AI Agents tab is disabled until an environment is selected; there’s no separate environment dropdown inside the tab itself
  2. Go to Settings → AI Agents and select the agent
  3. Turn off “Use Global” and select the model for that environment
Precedence order (highest wins):
An environment-scoped override on the agent itself wins over everything else; an org-scoped override on the agent wins over the environment and org defaults; and the Environment AI Defaults (when set) win over the Organization AI Defaults.

​
Available agents

sureops has exactly five agents that are individually configurable from Settings → AI Agents — the API rejects any agent ID outside this set: Other agents and background workflows (ECHO chat, SAGE incident chat, the Deep RCA Investigator, Fix-PR, the Incident Response Critic, the Improvement Recommender) run using the model resolved for their owning stage or the org default — they are not independently listed in this configuration screen today.
Pre-mortems (risk analysis on PRs before they ship) are on the roadmap and not yet available to configure here. See Pricing & Tiers for current roadmap status.

​
Available models

In BYOK mode, the dropdown reflects the models sureops has validated for agent use with your provider, not every model your provider offers. Model availability varies by plan. If you need a specific model that isn’t listed, contact hello@sureops.ai. Demo and trial organizations don’t get this dropdown at all — model selection is already fully locked today, the same way it will be for managed-mode organizations once that mode launches: sureops selects the model for each agent from a curated map, and only prompts, tasks, and skills stay customer-configurable. For details on managed vs. BYOK, see BYOK vs Managed LLM.

​
Cost considerations

Different models have different per-token costs. Agents running on high-capability models for every incident can meaningfully increase your LLM spend. A practical approach:
  • Set a cost-effective default for most agents
  • Override only the agents where output quality matters most (Diagnosis Specialist, Problem Commander)
  • Use per-environment overrides to use expensive models only in production
In BYOK mode, your org’s spend ceiling is visible at Settings → Budget (and the per-environment Budget tab). In demo/trial mode (and in managed mode once it launches), sureops owns the ceiling — you see a “budget managed by sureops” indicator rather than a self-serve, editable cap, since the spend is on sureops’s own provider accounts.

​
Budget and enforcement

Each org has a configurable LLM budget, tracked per stage and per incident across three dimensions: wall-clock time, number of LLM calls, and (BYOK) dollar cost. Budget enforcement has three modes: BYOK organizations default to Warn. Organizations in demo/trial mode (and managed organizations, once that mode launches) ship with Enforce by default, since sureops is the one absorbing the underlying provider cost. The enforce mode is designed for cost control, not safety. An enforced org that hits its ceiling has its incidents escalated to human review rather than silently failing.