PulseAugur
EN
LIVE 05:23:04

Marker 2 document converter achieves 5x throughput, beats competitors on benchmark

By PulseAugur Editorial · [3 sources] ·

Datalab has released Marker 2, a significantly rewritten open-source document conversion pipeline. The new version boasts a 5x increase in throughput compared to MinerU, achieving 2.9 pages per second on a single Nvidia B200 GPU while scoring 76.0% on the olmOCR-bench. Marker 2 was rebuilt with three core components, including Surya OCR 2 and a faster pdftext module, and offers multiple conversion paths optimized for quality, speed, or CPU-only operation. AI

IMPACT This release offers significant improvements in document processing speed and accuracy, potentially impacting workflows that rely on converting various file formats into structured data.

RANK_REASON Release of a new version of an open-source document conversion pipeline.

Read on MarkTechPost →

AI-generated summary · Google Gemini · from 3 sources. How we write summaries →

Marker 2 document converter achieves 5x throughput, beats competitors on benchmark

COVERAGE [3]

  1. MarkTechPost TIER_1 English(EN) · Asif Razzaq ·

    Datalab Marker v2 vs MinerU, Docling, and Liteparse: Benchmark Breakdown

    <p>Datalab rewrote Marker as a three-mode pipeline. Version 2 hits 76.0 on olmOCR-bench and sustains 2.9 pages per second on one B200 — over 5× MinerU's pipeline backend, while beating Docling on both accuracy and speed. Here's how it compares against MinerU, Docling and LitePars…

  2. MarkTechPost TIER_1 English(EN) · Asif Razzaq ·

    Marker 2 Beats MinerU, Docling and LiteParse with 5× Throughput

    <p>Datalab rewrote Marker as a three-mode pipeline. Version 2 hits 76.0 on olmOCR-bench and sustains 2.9 pages per second on one B200 — over 5× MinerU's pipeline backend, while beating Docling on both accuracy and speed. Here's how it compares against MinerU, Docling and LitePars…

  3. MarkTechPost TIER_1 English(EN) · Asif Razzaq ·

    Datalab’s Marker 2 vs MinerU, Docling and LiteParse: 76.0 on olmOCR-bench at 5× MinerU’s Throughput

    <p>Datalab rewrote Marker as a three-mode pipeline. Version 2 hits 76.0 on olmOCR-bench and sustains 2.9 pages per second on one B200 — over 5× MinerU's pipeline backend, while beating Docling on both accuracy and speed. Here's how it compares against MinerU, Docling and LitePars…

RELATED ENTITIES

RELATED TOPICS