Does anyone know if Vapi supports the "latency-optimized" configuration of Claude Haiku 3.5 via Bedrock? Here is the AWS docs page about it: https://docs.aws.amazon.com/bedrock/latest/userguide/latency-optimized-inference.html This would be a huge win for us if we could use it, since we have AWS credits and the TTFB is <400ms consistently.