llmboard.aiAI model intelligence
Home

Model Rankings

OverallOpen ModelsAgentCodingReasoningMathKnowledgeInstruction FollowingTextVision
Image GenerationImage Editing
Video GenerationImage to VideoVideo Editing
Text to SpeechSpeech to Text
Embeddings

Efficiency

Chat Token PricingImage PricingVideo PricingAudio Pricing
Chat Speed & LatencyProvider Reliability

Benchmarks

GPQAMMLU-ProAIME 2025SWE-Bench VerifiedMMLUHumanity's Last ExamLiveCodeBenchMATHHumanEvalMMMU-Pro
All Benchmarks

Tools

Model DirectoryCompare Models

Scoring & Data

Scoring & Data
393 models668 benchmarks

Leaderboard Center

Overall RankingCodingCore BenchmarksPrice & ValueRuntime Performance

Modalities

All ModelsImage GenerationImage EditingVideo GenerationImage-to-VideoVideo EditingText-to-SpeechSpeech-to-TextEmbeddings

Data & Methods

Scoring MethodAll BenchmarksReasoningMath

Vendors

All VendorsOpenAIAnthropicGoogle
llmboard.aiCopyright 2026 llmboard.ai

image to text benchmark

OCRBench-V2 (zh) Leaderboard

OCRBench v2 Chinese subset: Enhanced benchmark for evaluating Large Multimodal Models on visual text localization and reasoning with Chinese text content

Updated Aug 17, 2026

Models11
Model coverage11
MetricScore
EvidenceB

On this page

  • Ranking
  • Highlights
  • Distribution
  • Top models
  • About
  • FAQ

OCRBench-V2 (zh) Ranking

Higher score ranks better on this benchmark.

11 rows
Columns

Show columns

Sort by
Rank
Model
Score
Percentile
Participants
Evidence
Evaluated
Rank01ModelACQwen3 VL 235B A22B ThinkingAlibaba Cloud / Qwen TeamScore63.5%Percentile100.0%Participants11EvidenceCEvaluatedAug 17, 2026
Rank02ModelACQwen3 VL 32B ThinkingAlibaba Cloud / Qwen TeamScore62.1%Percentile90.0%Participants11EvidenceCEvaluatedAug 17, 2026
Rank03ModelACQwen3 VL 235B A22B InstructAlibaba Cloud / Qwen TeamScore61.8%Percentile80.0%Participants11EvidenceCEvaluatedAug 17, 2026
Rank04ModelACQwen3 VL 8B InstructAlibaba Cloud / Qwen TeamScore61.2%Percentile70.0%Participants11EvidenceCEvaluatedAug 17, 2026
Rank05ModelACQwen3 VL 30B A3B ThinkingAlibaba Cloud / Qwen TeamScore60.4%Percentile60.0%Participants11EvidenceCEvaluatedAug 17, 2026
Rank06ModelACQwen3 VL 32B InstructAlibaba Cloud / Qwen TeamScore59.2%Percentile50.0%Participants11EvidenceCEvaluatedAug 17, 2026
Rank07ModelACQwen3 VL 8B ThinkingAlibaba Cloud / Qwen TeamScore59.2%Percentile40.0%Participants11EvidenceCEvaluatedAug 17, 2026
Rank08ModelACQwen2.5 VL 32B InstructAlibaba Cloud / Qwen TeamScore59.1%Percentile30.0%Participants11EvidenceCEvaluatedAug 17, 2026
Rank09ModelACQwen3 VL 30B A3B InstructAlibaba Cloud / Qwen TeamScore57.8%Percentile20.0%Participants11EvidenceCEvaluatedAug 17, 2026
Rank10ModelACQwen3 VL 4B InstructAlibaba Cloud / Qwen TeamScore57.6%Percentile10.0%Participants11EvidenceCEvaluatedAug 17, 2026
Rank11ModelACQwen3 VL 4B ThinkingAlibaba Cloud / Qwen TeamScore55.8%Percentile0.0%Participants11EvidenceCEvaluatedAug 17, 2026

OCRBench-V2 (zh) Highlights

The leading models and scores on this benchmark.

Rank #1Qwen3 VL 235B A22B Thinking63.5%Rank #2Qwen3 VL 32B Thinking62.1%Rank #3Qwen3 VL 235B A22B Instruct61.8%Rank #4Qwen3 VL 8B Instruct61.2%

OCRBench-V2 (zh) Score Distribution

A closer view of the leading scores on this benchmark.

OCRBench-V2 (zh)

The Top AI Models for OCRBench-V2 (zh)

The first five results on this benchmark, with official price and output speed added where the model identity can be matched.

Ranking basisThis ocrbench-v2 (zh) AI model leaderboard uses descending score in the benchmark's original unit. The leaderboard ranking keeps matched price and speed data separate from benchmark evidence.

  1. 01
    AC
    Qwen3 VL 235B A22B ThinkingAlibaba Cloud / Qwen Team
    Score
    63.5%

    Strengths

    • Ranks #1 of 11 compared models
    • 100th percentile on this benchmark
    • C evidence result

    Considerations

    • This result measures OCRBench-V2 (zh), not total model capability
  2. 02
    AC
    Qwen3 VL 32B ThinkingAlibaba Cloud / Qwen Team
    Score
    62.1%

    Strengths

    • Ranks #2 of 11 compared models
    • 90th percentile on this benchmark
    • C evidence result

    Considerations

    • This result measures OCRBench-V2 (zh), not total model capability
  3. 03
    AC
    Qwen3 VL 235B A22B InstructAlibaba Cloud / Qwen Team
    Score
    61.8%

    Strengths

    • Ranks #3 of 11 compared models
    • 80th percentile on this benchmark
    • C evidence result

    Considerations

    • This result measures OCRBench-V2 (zh), not total model capability
  4. 04
    AC
    Qwen3 VL 8B InstructAlibaba Cloud / Qwen Team
    Score
    61.2%

    Strengths

    • Ranks #4 of 11 compared models
    • 70th percentile on this benchmark
    • C evidence result

    Considerations

    • This result measures OCRBench-V2 (zh), not total model capability
  5. 05
    AC
    Qwen3 VL 30B A3B ThinkingAlibaba Cloud / Qwen Team
    Score
    60.4%

    Strengths

    • Ranks #5 of 11 compared models
    • 60th percentile on this benchmark
    • C evidence result

    Considerations

    • This result measures OCRBench-V2 (zh), not total model capability

Selection summary

Best AI Models for OCRBench-V2 (zh)

Qwen3 VL 235B A22B Thinking currently leads OCRBench-V2 (zh) with 63.5%. It is the top model on this specific benchmark, while the best LLM for the broader task should also be checked against other benchmarks, price and runtime.

Use this leaderboard with the supporting benchmark results and coverage details above. A leaderboard position summarizes the selected ranking signal; it does not replace workload-specific testing.

Benchmark rank #1Qwen3 VL 235B A22B Thinking63.5%Benchmark rank #2Qwen3 VL 32B Thinking62.1%Benchmark rank #3Qwen3 VL 235B A22B Instruct61.8%

What is OCRBench-V2 (zh)?

What OCRBench-V2 (zh) measures and how its scores work.

OCRBench v2 Chinese subset: Enhanced benchmark for evaluating Large Multimodal Models on visual text localization and reasoning with Chinese text content

Scores are shown in ratio. This benchmark is not independently verified and has an evidence level of B.

Family
OCRBench-V2 (zh)
Modality
multimodal
Primary category
image to text
Score direction
higher
LLMBoard eligible
Yes
Evaluation key
ocrbench-v2-(zh)|llm-stats-current

Benchmark scores retain their original unit. Overall score eligibility is shown separately.

FAQ

Common questions about OCRBench-V2 (zh).

Which model scores highest on OCRBench-V2 (zh)?

Qwen3 VL 235B A22B Thinking is currently ranked first with 63.5%.

What does OCRBench-V2 (zh) measure?

OCRBench v2 Chinese subset: Enhanced benchmark for evaluating Large Multimodal Models on visual text localization and reasoning with Chinese text content

Is a higher score better?

Yes. Higher values rank better for this benchmark.

How many models are compared?

11 model results are currently shown.

Does this benchmark affect the overall score?

Yes. This benchmark can contribute to the current LLMBoard capability score.