模型多源确认

GPT-6.1 Sol在文档OCR任务中表现提升

精选理由

GPT-6.1 Sol在文档OCR方面有显著进步,特别是表格解析能力提升明显,与GPT-6 Astra相当。

研究人员对GPT-6.1 Sol进行了全面的文档OCR基准测试。与一周前的GPT-6 Sol相比,GPT-6.1 Sol在表格解析能力和阅读顺序方面有显著提升。GPT-6.1 Sol的表格解析能力与GPT-6 Astra相当。ParseBench平台提供了全面的文档解析能力评估。

原文 · Jerry Liu

we comprehensively benchmarked gpt-6.1 sol on document OCR tasks. compared to gpt-6 sol from a week ago (!), we see a sizable increase in table parsing capabilities and reading order, which is pleasantly surprising. gpt-6.1 sol has similar table parsing as gpt-6 astra. these are great results. frontier models are still typically an order of magnitude more expensive than cost-effective document parsing solutions, but this also means that there's a lot of alpha in improving capabilities at the "premium" end (>1c per page) of the spectrum too check out ParseBench: parsebench.ai for a comprehensive set of document parsing capabilities, check out llamaparse: cloud.llamaindex.ai 💬 1 🔄 0 ❤️ 2 👀 391 📊 2 ⚡

  • Tanishq Abraham (论文推介)09-29 17:23原文
  • IT之家09-28 01:10原文
  • Perplexity09-28 15:10原文
  • kimmonismus09-28 18:28原文
  • lmarena.ai09-28 20:44原文
  • PolymarketMoney09-28 22:10原文
  • Andrew Curran09-28 23:15原文
  • Gary Marcus09-29 00:35原文
  • 宝玉09-29 01:15原文
  • The Business Times: Tech09-29 01:23原文