OCR Bench UFO

Pairwise VLM-as-judge OCR benchmark ranking OCR models on UFO document images with Bradley-Terry ELO scores.

4rows
eloprimary metric
2026-05-06sampled

Metadata

ID
ocr_bench_ufo
Category
Multimodal
Release
Unknown

Metrics

ELO, Win%, Wins, Losses (lower is better), Ties, ELO Low, ELO High

Latest Results

Rows are parsed from the public Hugging Face dataset-server rows API. ELO rankings come from OCR Bench pairwise VLM-as-judge comparisons.

Rank Subject ELO Model Match Provenance Sampled
1 deepseek-ai/DeepSeek-OCR 1691 — Imported 2026-05-06
2 lightonai/LightOnOCR-2-1B 1570 — Imported 2026-05-06
3 rednote-hilab/dots.ocr 1432 — Imported 2026-05-06
4 zai-org/GLM-OCR 1307 — Imported 2026-05-06