Ask HN: Slow OpenAI Inference on AWS Bedrock

Inference on AWS Bedrock for OpenAI models has been very slow and glitchy the last few days with today being the worst. Very long latency of over 5 minutes to first token. What's going on?

3 points | by timedude 2 days ago

2 comments

  • mikert89 1 day ago
    likely they sold all the gpu capacity to enterprise clients
  • timedude 2 days ago
    us-east-1 seems to be the worst region right now.