AI Model

1. Model Selection

A key piece of building an effective agent is choosing the right foundation AI model for your use case.
By default, Credal gives you access to the latest set of models from major providers: OpenAI, Anthropic and Google.

Here you can choose the foundation AI model you want your agent to use from the drop-down menu.
More advanced models, like o1, are both more expensive and more capable than earlier models.
When selecting a model, you may want to consider:

• Reasoning capabilities: Different models vary in their ability to execute multi-step reasoning, decompose complex tasks and follow nuanced instructions.
• Domain-specific expertise: Strengths can vary across domains like code generation, creative writing, analysis & research and image processing.
• Context window sizes: Models with larger context windows can process more input (prompts, data sources, codebases) and sustain longer conversations.
• Tool calling limitations: This includes the ability to select the right tools, extract/format requests, make parallel requests and handle responses (objects, streaming, etc.).
• Speed & cost: More capable models may be slower and more expensive. Response time and budget constraints should guide your choice.

Tip: GPT-4o is a good starting point for most use cases, as it’s very capable and relatively affordable. If you’re unsure, run test queries to gauge performance for your needs.

Credal can help you make an informed choice with:

• Short descriptions of each model covering the above factors
• A “Preview” tab to test agent responses before deploying
• Recommendations from our team — just ask, we’re happy to help

If there’s a model you’d like to use but don’t see in the platform, reach out so we can look into adding it.

---

2. Creativity

You can set the level of creativity you want your agent to have by selecting from the panel below.
For most use cases, “precise” or “balanced” is recommended.

• Higher creativity → The agent may produce unusual, unexpected responses, great for brainstorming, marketing and creative writing.
• Higher precision → The agent relies more on the context and data you provide, better for analysis, research and tasks requiring accuracy.

Think of the creativity setting like having an “eccentric genius” on the team:
A lot of what it says may be off, but occasionally it produces something uniquely valuable.