[Question]: Is it normal for documents to take a while to process?

Do you need to ask a question?

• I have searched the existing question and discussions and this question is not already answered.
• I believe this is a legitimate question, not just a bug or feature request.

Your Question

Hey there, I have some pretty extreme hardware,

CPU: 9800x3d GPU: 5090

Yet, doing something as simple as processing a couple of ~12-page PDFs with a normal amount of images takes 15 minutes... Do you know if this is normal?

My GPU is not being utilised at all, and my CPU has very minimal utilisation...

It seems the MinerU command executed successfully happens quickly, but after this, the processing is super slow.

[07/09/25 00:52:04] INFO     Creating llm function with model: gpt-4o-mini                                                                                                                                                                                                              main.py:55
                    INFO     Creating embedding function with model: text-embedding-3-large                                                                                                                                                                                             main.py:93
INFO: Process 28784 Shared-Data created for Single Process
                    INFO     Init {'embedding_dim': 3072, 'metric': 'cosine', 'storage_file': '/shared_workspace/vdb_entities.json'} 0 data                                                                                                                     dbs.py:81
                    INFO     Init {'embedding_dim': 3072, 'metric': 'cosine', 'storage_file': '/shared_workspace/vdb_relationships.json'} 0 data                                                                                                                dbs.py:81
                    INFO     Init {'embedding_dim': 3072, 'metric': 'cosine', 'storage_file': '/shared_workspace/vdb_chunks.json'} 0 data                                                                                                                       dbs.py:81
INFO: Process 28784 initialized updated flags for namespace: [full_docs]
INFO: Process 28784 ready to initialize storage namespace: [full_docs]
INFO: Process 28784 initialized updated flags for namespace: [text_chunks]
INFO: Process 28784 ready to initialize storage namespace: [text_chunks]
INFO: Process 28784 initialized updated flags for namespace: [entities]
INFO: Process 28784 initialized updated flags for namespace: [relationships]
INFO: Process 28784 initialized updated flags for namespace: [chunks]
INFO: Process 28784 initialized updated flags for namespace: [chunk_entity_relation]
INFO: Process 28784 initialized updated flags for namespace: [llm_response_cache]
INFO: Process 28784 ready to initialize storage namespace: [llm_response_cache]
INFO: Process 28784 initialized updated flags for namespace: [doc_status]
INFO: Process 28784 ready to initialize storage namespace: [doc_status]
INFO: Process 28784 storage namespace already initialized: [full_docs]
INFO: Process 28784 storage namespace already initialized: [text_chunks]
INFO: Process 28784 storage namespace already initialized: [llm_response_cache]
INFO: Process 28784 storage namespace already initialized: [doc_status]
INFO: Process 28784 Pipeline namespace initialized
MinerU command executed successfully
MinerU command executed successfully
Error parsing equation analysis response: Invalid \escape: line 2 column 689 (char 690)
Error parsing equation analysis response: Invalid \escape: line 2 column 62 (char 63)
Error parsing equation analysis response: Invalid \escape: line 2 column 262 (char 263)
[07/09/25 01:04:39] INFO     Creating llm function with model: gpt-4o-mini 
Indexed 2 new or updated file(s) into shared workspace '/home/jmerh/.rag_anything/shared_workspace'.

Is there some sort of config I need to change?

Additional Context

No response