Briefs

BriefsWorkflowAvailable now1 min read

OpenRouter published a guide on pulling tables from PDFs with LLMs

The tutorial outlines how to pass PDFs directly to LLMs with a strict JSON schema per row and catch silent parser failures.

OpenRouter published a guide on pulling tables from PDFs with LLMs — image from OpenRouter
Image: OpenRouter

Summary

OpenRouter published a guide on pulling clean table data from PDFs by passing files directly to an LLM with a strict JSON schema for one row. In benchmarks across Claude Haiku 5.5, GPT-6 Luna, and Gemini 3.8 Flash, direct file reading got all 644 cells right. The workflow includes guardrails against silent failures and token limits.

What changed

  • Send PDFs straight to file-reading LLMs with a strict JSON schema describing a single row to get typed table outputs.
  • Every model tested that read the PDF directly—including Claude Haiku 5.5, GPT-6 Luna, and Gemini 3.8 Flash—got all 644 cells right.
  • Reject responses where finish_reason is length to catch truncated tables, and reject empty outputs from silent parser failures.
  • Optionally set engine='mistral-ocr' to run OCR on each page before processing for $2 per 1,000 pages on top of token costs.

Availability

Available now

“To extract tables from a PDF with an LLM, send the PDF to a model that can read files , together with a strict JSON schema that describes one row of the table.”

Founder takeaway

If you parse complex PDFs, test sending the document directly to a file-reading LLM alongside a strict row schema. Add explicit checks for token cutoffs and zero-row parser returns.

Source

OpenRouterHow to Extract Tables from a PDF with an LLM (Python)

Announced Checked

Researched and drafted with AI from the official sources above, then checked automatically before publishing. Spotted a mistake? Tell us.

SharePost on XLinkedIn

Related

More briefs

All briefs
  1. LangChain open-sources its AI paid media agent — image from LangChain

    WorkflowAvailable now

    LangChain open-sources its AI paid media agent

    LangChain details the architecture and technical lessons behind its open-source paid media agent that manages advertising campaigns.

    LangChain1 min read

  2. Supabase adds Make integration for AI agents and workflows — image from Supabase

    WorkflowAvailable now

    Supabase adds Make integration for AI agents and workflows

    Connect your Supabase account to Make to let automation workflows and AI agents query databases, read logs, and deploy Edge Functions.

    Supabase1 min read

  3. AWS details distributed PyTorch training with Ray Train DLCs and EFA on EKS — image from AWS

    WorkflowAvailable now

    AWS details distributed PyTorch training with Ray Train DLCs and EFA on EKS

    AWS published a guide on running distributed PyTorch training on Amazon EKS using Ray Train Deep Learning Containers and Elastic Fabric Adapter.

    AWS1 min read

One email, every Monday

Briefs are published here through the day. Monday Heist is the email: one useful AI workflow, the files to use it, and the catch. The first issue is being prepared; subscribers get it first. Free, and you can leave in one click.

One email. Every Monday.