#Slide shows?

1 messages · Page 1 of 1 (latest)

royal briar
#

Can the Assistants API research a subject and then create a slide show?

boreal pollenBOT
#

Hi @royal briar. I'm kapa.ai, your friendly support bot. I will try to answer your question by searching through the knowledge base. :wave:

#

Yes, the Assistants API can help automate the creation of slide shows, including researching a subject and generating slides with content and visuals.

How it works:

  1. Researching a Subject:
    The Assistants API can be configured to act as a research assistant, using advanced models (like GPT-4 or o-series models) to analyze data, extract insights, and summarize findings. You can provide instructions for the assistant to research a topic, analyze data, and generate key insights or summaries. For more complex research, you can use the deep research models with the Responses API, but the Assistants API itself can handle many research and summarization tasks as shown in the slide creation example Creating slides with the Assistants API and DALL·E 3.

  2. Creating Slide Content:
    The API can generate text for slide titles, bullet points, and summaries. It can also analyze uploaded data files (e.g., CSV or JSON), perform calculations, and create visualizations (like charts or plots) using the Code Interpreter tool. The assistant can then summarize the findings and generate concise slide content see example.

  3. Generating Visuals:
    You can use DALL·E 3 to generate custom images for your slides. While DALL·E 3 is not yet natively integrated into the Assistants API, you can generate images separately and upload them for use in your slides see example.

#
  1. Building the Slide Deck:
    The assistant can use code templates (e.g., with the python-pptx library) to assemble the researched content, generated images, and visualizations into a PowerPoint (.pptx) file. The process can be fully automated, from research to slide generation, as demonstrated in the OpenAI Cookbook example see full workflow.

Summary:
With the Assistants API, you can automate the end-to-end process of researching a topic and creating a slide show, including data analysis, insight generation, image creation, and slide assembly. This can save significant time and effort compared to manual slide creation.

For a detailed, step-by-step example, see the OpenAI Cookbook: Creating slides with the Assistants API and DALL·E 3.

Note: you can ask me a follow up question by @ mentioning me again :speech_balloon:

#

@royal briar Kapa.ai is still learning and improving, please let me know how I did by reacting below.

royal briar
#

@boreal pollen - How much do you think it would cost to use the above mentioned automation to create slide decks of around 26 slides (including visuals).

boreal pollenBOT
#

Let me check the knowledge base for you again. One moment! :books:

#

The cost to automate the creation of a 26-slide deck (including visuals) using the Assistants API and DALL·E 3 depends on several factors:

1. Image Generation (DALL·E 3)

  • DALL·E 3 Pricing:
    • 1024x1024, Standard: $0.04 per image
    • 1024x1024, HD: $0.08 per image
    • Larger sizes (1024x1536 or 1536x1024): $0.08 (Standard), $0.12 (HD) per image
      See pricing details
  • Example Calculation:
    If you generate 1 image per slide (26 images, 1024x1024, Standard):
    26 × $0.04 = $1.04
    If you use HD quality:
    26 × $0.08 = $2.08

2. Text Generation (Assistants API)

  • Model Pricing:
    • GPT-4o: $5.00 per 1M input tokens, $15.00 per 1M output tokens
    • GPT-4 Turbo: $10.00 per 1M input tokens, $30.00 per 1M output tokens
      See model pricing
  • Token Usage Estimate:
    • A typical slide deck (text only) is unlikely to exceed a few thousand tokens. For example, 10,000 tokens (input + output) would cost less than $0.20 with GPT-4o.

3. Code Interpreter Sessions

4. File Storage (if using File Search)

  • Pricing:
    • $0.10/GB/day (first GB free)
    • Not typically significant for a single slide deck unless storing large files.

Estimated Total Cost (per 26-slide deck):

  • Images (Standard quality): ~$1.04
  • Text generation: Likely <$0.20
  • Code Interpreter: $0.03
  • File Storage: Usually $0 (if under 1GB)
#

Total: ~$1.27 (Standard quality images) to ~$2.31 (HD images), plus a small amount for text generation and file storage if applicable.

Note: These are rough estimates. Actual costs may vary depending on the amount of text, number of images, and any additional API usage. For precise budgeting, monitor your usage in the OpenAI dashboard and refer to the official pricing page Managing costs.

If you need more detailed cost modeling, you can use the tokenizer tool to estimate token usage for your specific content.

Note: you can ask me a follow up question by @ mentioning me again :speech_balloon:

#

@royal briar Kapa.ai is still learning and improving, please let me know how I did by reacting below.