Skip to content
Trending storyDeveloping

GPT 6.1 Terra ranks fifth on LlamaIndex's ParseBench document parsing leaderboard

1 article1 sourcesince Oct 27Last article

Overview

AISummary of 1 article

Jerry Liu of LlamaIndex reports that GPT 6.1 Terra ranks fifth overall on the ParseBench document parsing leaderboard, ahead of 6 Sol and behind Opus 5.5 and Astra.

Liu says the model performs especially well at table parsing and costs about 1 cent per page.

Liu also says the model is decent at content faithfulness and reading order but struggles with layout, charts, and formatting. He suggests pairing it with specialized OCR tools such as LlamaParse for more consistent results.

Written by AI from the articles below · updated Oct 27, 9:28 PM ET

Check the sources:

Latest developmentOct 27, 8:57 PM ET
GPT 6.1 Terra ranks fifth on ParseBench for table parsing

Article timeline

The articles in this story. Times are ET.

Oct 27
  1. Jerry LiuX
    GPT 6.1 Terra ranks fifth on ParseBench for table parsing

    AILlamaIndex reports that GPT 6.1 Terra ranks fifth on its ParseBench leaderboard, with strong table parsing, ahead of 6 Sol and behind Opus 5.5 and Astra. The model is decent at content faithfulness and reading order but struggles with layout, charts, and formatting. LlamaIndex says it costs about 1 cent per page and suggests pairing it with specialized OCR tools such as LlamaParse for more consistent results.

    Image from @jerryjliu0's post

Heat Trend

Current comparable heat
16
Peak comparable heat
20Oct 27, 10:00 PM ET
Change in past 24 hours
–

The trend compares only continuously observed participants, so its scope may be smaller than the current heat total. Hover or tap for hourly heat; use arrow keys with a keyboard.