Versions Compared

Key

  • This line was added.
  • This line was removed.
  • Formatting was changed.

Gateway Vector DB

The gateway stores the text of the content uploaded.
Assuming a five-char average per word and 500 words per page, it is estimated that 1 page of content should take 5kb.
The system stores the original text and a vector representation (chunk) of the content, approximately 2kb per page.

...

 

Site A

Site B

Word                     No. of Files 

                               Size  

 Average word file size

1214  

240 MB 

200Kb

480 

400 MB

800Kb 

Excel                     No. of Files 

                              Size 

 

283  

37 MB 

281 

24 MB 

 

PowerPoint          No. of Files 

                               Size  

 

472  

2368 MB 

27  

74 MB 

Pdf                        No. of Files  

                              Size  

 Average file size

1372  

2262 MB 

1600 Kb

975 

407 MB 

400Kb

Number of Documents chanks

Each Chunk is 500 tokens.

...

If the site has 150,000 chunks - it contains 150,000*0.75= 112.500 pages, which is around 10K documents.

Memory usage for the Postgres DB

it is recommended to run with at least 8G for the postgress. Even when empty, it uses 4G.

For 1M chunks, it takes around 8G ram (around 350,000 documents according to AGAT)

Gateway SQL server BD sizing

It is recommended to that the SQL Server run with at least 4GB.

Cost estimations for using OpenAI services

Embeddings

Assuming a typical text page is 500 words, calculate the cost of embedding a typical page; let's break it down:

  1. Estimate the number of tokens per page:

    • You mentioned that 100 tokens are around 75 words.

    • A typical page is around 500 to 600 words, translating to approximately 700 tokens.

  2. Determine the cost per token:

    • The cost for embeddings (Text Embedding 3 small) is $0.02 for 1 million tokens.

  3. Calculate the cost for

...

  1. one page:

    • Since 700 tokens are on a typical page, we can calculate the cost as follows:

    Cost per page=700 tokens1,000,000 tokens×0.02 USD\text{Cost per page} = \frac{700 \text{ tokens}}{1,000,000 \text{ tokens}} \times 0.02 \text{ USD} Cost per page=1,000,000 tokens700 tokens​×0.02 USD Cost per page=0.000014 USD\text{Cost per page} = 0.000014 \text{ USD}Cost per page=0.000014 USD

So, the cost for embedding a typical page (approximately 700 tokens) would be per page.

Or in other words - 1$ can produce 70K pages

Questions