Python couldn't get the same results as those from web UI's preview tab #202

Closed
opened 2026-02-16 00:17:07 -05:00 by yindo · 1 comment
Owner

Originally created by @stonesthatwhisper on GitHub (Jul 24, 2024).

Describe the bug
Python LlamaParse is not getting the same results as those I got from the web UI preview.

parser = LlamaParse(
    api_key=...
    result_type="markdown",  # "markdown" and "text" are available
    verbose=True,
    language="ch_sim",
    use_vendor_multimodal_model=True,
    vendor_multimodal_model_name="openai-gpt4o",
    vendor_multimodal_api_key=os.environ['OPENAI_API_KEY'],
)

is getting different results from those in the UI's preview tab, which are
more or less satisfactory. The result I got from Python SDK seems not
getting use_vendor_multimodal_model=True, and also not respecting
the language argument I set too.

Job ID
cac11eca-1350-45e9-889b-a844a962d003

Screenshots
Screen Shot 2024-07-24 at 11 19 31 PM

Client:
Please remove untested options:

  • Frontend (cloud.llamaindex.ai)
  • Python Library

Options
Multimodal

Originally created by @stonesthatwhisper on GitHub (Jul 24, 2024). **Describe the bug** Python LlamaParse is not getting the same results as those I got from the web UI preview. ``` parser = LlamaParse( api_key=... result_type="markdown", # "markdown" and "text" are available verbose=True, language="ch_sim", use_vendor_multimodal_model=True, vendor_multimodal_model_name="openai-gpt4o", vendor_multimodal_api_key=os.environ['OPENAI_API_KEY'], ) ``` is getting different results from those in the UI's preview tab, which are more or less satisfactory. The result I got from Python SDK seems not getting use_vendor_multimodal_model=True, and also not respecting the language argument I set too. **Job ID** cac11eca-1350-45e9-889b-a844a962d003 **Screenshots** <img width="1112" alt="Screen Shot 2024-07-24 at 11 19 31 PM" src="https://github.com/user-attachments/assets/d90f4d9b-3a62-47b0-b3bb-2e0fc5e3f80e"> **Client:** Please remove untested options: - Frontend (cloud.llamaindex.ai) - Python Library **Options** Multimodal
yindo added the bug label 2026-02-16 00:17:07 -05:00
yindo closed this issue 2026-02-16 00:17:07 -05:00
Author
Owner

@stonesthatwhisper commented on GitHub (Jul 24, 2024):

I am running these in a Jupyter-lab notebook. After restarting the kernel and rerunning every step. I am getting the expected results. Closing this issue.

@stonesthatwhisper commented on GitHub (Jul 24, 2024): I am running these in a Jupyter-lab notebook. After restarting the kernel and rerunning every step. I am getting the expected results. Closing this issue.
Sign in to join this conversation.
1 Participants
Notifications
Due Date
No due date set.
Dependencies

No dependencies set.

Reference: run-llama/llama_cloud_services#202