awslabs / agent-evaluation

A generative AI-powered framework for testing virtual agents.
https://awslabs.github.io/agent-evaluation/
Apache License 2.0
64 stars 10 forks source link

Add support for Claude 3 Haiku #13

Open tonykchen opened 2 months ago

tonykchen commented 2 months ago

Add Claude 3 Haiku as an available model under ClaudeEvaluator

tonykchen commented 2 months ago

Claude 3 Haiku has been enabled as part of #28, but we still need to optimize the prompts to get consistent results. It's likely we'll need a different set of prompts specific for Haiku.