Agent Gallery
Agent Gallery OperationsRunning Agents

Example: Running a Multi-Model Agent

Configure the Multi LLM Cross-Check Server with multiple provider keys and validate responses.

Before You Start

Login required

Running agents (clicking the run button) requires you to be logged in to the gallery. Refer to Logging In to Agent Gallery if you need to authenticate.

  • Log in to Agent Gallery with your account.
  • Gather API keys for each model you plan to compare (for example, Claude, Perplexity, Gemini).

Launch Steps

  1. In Agent Gallery, search for multi and open the Multi LLM Cross-Check Server agent page. Multi LLM Cross-Check Server detail page with the Run button highlighted
  2. Click Run, expand Optional Settings, and paste the API keys for every provider you want to cross-check. Run Agent dialog showing API key fields for multiple models
  3. Select Connect, wait for the status indicator to show Agent is Running, then choose Go to Agent to open the live console.
  4. Confirm the agent is Running before launching validation prompts.

Validation Queries

Run cross-check prompts to ensure every model responds and differences appear clearly:

cross check models on the query: what is accenture's ai refinery's main selling point. be concise and sharp. present outputs of the cross check as a markdown table.

Multi-model comparison table for the AI Refinery prompt

Executing a time-sensitive prompt, the day after the F1 night race was held in Singapore. Only perplexity, which by default does a web search first, gets the answer correct. The other models all have a knowledge cut-off.

did lando win the f1 singapore race last night? present outputs of the cross check as a markdown table.

Cross-check results

Shut Down the Agent

When validation is complete, choose Shutdown Agent from the run console (beside the Running status) to release resources.