VECTOR WIREAI INTELLIGENCE
PKT
Refresh Models Deals Regulatory Sources

Questions this page answers

Which model leads CC-OCR?
Across 24 models scored on CC-OCR, Qwen3.6 Plus leads at 83.4, ahead of Qwen3 VL 235B A22B Instruct at 82.2. The median tracked score is 79.8, and the field spans 68.1 to 83.4.

CC-OCR

Across 24 models scored on CC-OCR, Qwen3.6 Plus leads at 83.4, ahead of Qwen3 VL 235B A22B Instruct at 82.2. The median tracked score is 79.8, and the field spans 68.1 to 83.4.

24 models tracked
Data as of August 25, 2026
#ModelVendorBest scoreRunsLast seen
1Qwen3.6 PlusAlibaba83.412026-08-25
2Qwen3 VL 235B A22B InstructAlibaba82.212026-08-25
3Qwen3.5 397B A17BAlibaba8212026-08-24
4Qwen3.6 35B A3BAlibaba81.922026-08-25
5Qwen3.5 122B A10BAlibaba81.812026-08-25
6Qwen3 VL 235B A22B ReasoningAlibaba81.512026-08-25
7Qwen3.6 27BAlibaba81.222026-08-25
8Qwen3.5 27BAlibaba8132026-08-25
9Qwen3.5 35B A3BAlibaba80.722026-08-25
10Qwen3 VL 30B A3B InstructAlibaba80.712026-08-25
11Qwen3 VL 32B InstructAlibaba80.312026-08-25
12Qwen3 VL 8B InstructAlibaba79.912026-08-25
13Qwen2.5 VL 72BAlibaba79.812026-08-25
14Qwen3 VL 30B A3B ReasoningAlibaba77.812026-08-25
15Qwen2.5 VL 32B InstructAlibaba77.112026-08-25
16Qwen2.5 VL 32BAlibaba77.112026-08-25
17Claude Opus 4.5Anthropic76.912026-08-24
18Qwen3 VL Thinking (8B)Alibaba76.312026-08-25
19Qwen3 VL 4B InstructAlibaba76.212026-08-25
20Gemma 4 31BGoogle75.722026-08-24
21Gemma 4 26B A4BGoogle74.512026-08-24
22Qwen3 VL 4B (Reasoning)Alibaba73.812026-08-25
23Qwen2 VL 72B InstructAlibaba68.712026-08-25
24Claude Sonnet 4.5Anthropic68.112026-08-24

Best tracked score per model (default configuration; source-attributed and verification-tiered). Open a model for its full benchmark surface, provenance and pricing.