Model Selection

Every agent needs a brain. On Abundly, you can choose which large language model (LLM) powers your agent—or simply let the platform pick the best one for you.

​

Default behavior

If you don’t have a preference, leave the model set to (no preference). This is the default, and it means Abundly will use whatever model we think works best for general agentic behavior. We continuously evaluate models and update this default as better options become available. This is the right choice for most users—you get great performance without having to think about model selection at all.

​

Model aliases vs specific versions

When selecting a model, you can choose between:

• Aliases like “Claude Sonnet Latest” — automatically points to the latest version that we have verified works well
• Specific versions like “Claude Sonnet 4.5” — locked to that exact version

We recommend using aliases in most cases. Models improve constantly, and when a new version of Claude Sonnet arrives, it usually makes sense to upgrade. By selecting an alias, you won’t need to remember to do that—Abundly will point the alias to the latest version when we’ve verified it works well (typically after some internal testing). However, if you want full control and want to minimize the risk of surprises, pick a specific model version. Although newer models usually perform better, agent behavior can sometimes change subtly. Locking to a specific version lets you manually decide when to upgrade.

​

Available models

Model availability is workspace-specific. Workspace admins can turn off specific catalog models for the entire workspace from Workspace management → Model selection—useful for controlling costs or standardizing on a smaller set of models. Models that Abundly has disabled platform-wide remain locked and cannot be re-enabled at the workspace level.

​

Policy for new models

When Abundly adds a new model to the catalog, your workspace can either make it available automatically or hold it back until an admin reviews it. Choose the policy under Workspace management → Settings → LLM models:

• New models on by default (the default) — New catalog models become available to your agents as soon as Abundly adds them. You can still turn any model off afterwards on the Model selection tab.
• Require approval for new models — New catalog models stay off until an admin enables them on the Model selection tab. Switching to this mode does not disable any model that is currently available; only models added to the catalog after you switch require an explicit enable.

Most workspaces use a mix of:

• Anthropic Claude
• OpenAI GPT
• Google Gemini
• xAI Grok

Some workspaces also include additional families such as Qwen (EU-hosted via Scaleway serverless inference), DeepSeek, or self-hosted models. You can switch models at any time—even mid-conversation.

​

Comparing models

Not sure which model to choose? We tend to default to the Claude models (Sonnet or Opus). We find they work very well for agentic behaviour. But the other models have improved a lot lately, and some have specific strenghts that you may want to leverage. This is a changing landscape, so you’re probably best off researching online yourself (or asking an Agent to do it for you…). But here’s a high level overview of how the models from each provider compare against each other.

​

Anthropic Claude models

​

Google Gemini models

​

OpenAI GPT models

​

xAI Grok models

​

Per-context model selection

Different contexts in your agent’s life may benefit from different models. You can configure a model and thinking preference per context:

• Chat — Conversations you start with the agent
• Scheduled tasks — When the agent runs on a schedule
• Email — When the agent responds to incoming emails
• Slack — When the agent responds to Slack messages
• Microsoft Teams — When the agent responds to Teams messages
• Agent messages — When other agents send your agent a message

If a context has no override, it inherits the agent’s default model. This lets you use a fast, cheap model for routine scheduled work while keeping a more capable model for interactive chat.

​

Per-task model overrides

Individual scheduled tasks can have their own model and thinking setting on top of the per-context default. Open a task’s settings card to configure it. This is useful when one particular task needs heavier reasoning than your other scheduled tasks.

​

Thinking mode

Many models support a thinking mode where the model spends more effort reasoning before it responds. Thinking makes your agent smarter at complex reasoning tasks, but it also increases processing time and credit usage. You can configure thinking in the default model settings, per context, and on individual scheduled tasks.

​

When to change models

​

Unified interface

Regardless of which model you choose, the platform provides a consistent experience:

• Switch models even mid-conversation
• Same capabilities work across all models
• Consistent behavior and tool usage
• No need to learn different APIs

​

Learn more