Skip to main content

Benchmark profile

OCRBench V2

A native OCR benchmark for reading text from images across multilingual scripts, low-quality scans, handwriting, structured layouts, charts, and screenshots.

Data verified

Benchmark score on OCRBench V2 — July 23, 2026

BenchLM mirrors the published score view for OCRBench V2. Qwen3.7 Plus leads the public snapshot at 70.7% , followed by Interfaze Beta (70.7%). BenchLM does not use these results to rank models overall.

2 modelsMultimodal & GroundedCurrentDisplay onlyUpdated July 23, 2026

Benchmark score table (2 models)

Score
1
Qwen3.7 PlusAlibaba · Closed
70.7%
2
Interfaze BetaInterfaze · Closed
70.7%

About OCRBench V2

Year

2025

Tasks

Image OCR tasks

Format

Accuracy

Difficulty

Native visual text understanding

OCRBench V2 evaluates whether multimodal models can extract visual text directly from images before downstream reasoning or structure extraction. BenchLM stores Interfaze's reported score as a display-only OCR row.

BenchLM freshness & provenance

Version

OCRBench V2 2025

Refresh cadence

Quarterly

Staleness state

Current

Question availability

Public benchmark set

CurrentDisplay only

BenchLM uses freshness metadata to decide whether a benchmark should still be treated as a strong differentiator, a benchmark to watch, or a display-only reference. For the full scoring policy, see the BenchLM methodology page.

FAQ

What does OCRBench V2 measure?

A native OCR benchmark for reading text from images across multilingual scripts, low-quality scans, handwriting, structured layouts, charts, and screenshots.

Which model scores highest on OCRBench V2?

Qwen3.7 Plus by Alibaba currently leads with a score of 70.7% on OCRBench V2.

How many models are evaluated on OCRBench V2?

2 AI models have been evaluated on OCRBench V2 on BenchLM.

Compare Top Models on OCRBench V2

Last updated: July 23, 2026 · BenchLM version OCRBench V2 2025

Choose a model with this week’s evidence

Join 2,000+ readers for ranking moves, pricing changes, and the claims that still need proof.

One email each week. Unsubscribe anytime.