The post indicates that the llama index is using client version 1.7.3. In the comments, a community member suggests that the latest client version 1.8 should be working. Another community member mentions they will need to check the project they are using, as it might be using an older version. The third comment provides an answer, stating that there was a max_opimization_threads set to null in qdrant by default, which the llama_index did not like, and setting it to 5 resolved the issue.
