CC-OCR
Qwen3.5-Omni-Plus 83.4% leads since Mar 2026 Fading Models read text in 7,058 images across four tracks: scene text, multilingual text, document parsing, and key information extraction. Each track has its own metric.
In the authors’ words
Large Multimodal Models (LMMs) have demonstrated impressive performance in recognizing document images with natural language instructions. However, it remains unclear to what extent capabilities in literacy with rich structure and fine-grained visual challenges. The current landscape lacks a comprehensive benchmark to effectively measure the literate capabilities of LMMs. Existing benchmarks are often limited by narrow scenarios and specified tasks. To this end, we introduce CC-OCR, a comprehensive benchmark that possesses a diverse range of scenarios, tasks, and challenges. CC-OCR comprises four OCR-centric tracks: multi-scene text reading, multilingual text reading, document parsing, and key information extraction. It includes 39 subsets with 7,058 full annotated images, of which 41% are sourced from real applications, and released for the first time. We evaluate nine prominent LMMs and reveal both the strengths and weaknesses of these models, particularly in text grounding, multi-orientation, and hallucination of repetition. CC-OCR aims to comprehensively evaluate the capabilities of LMMs on OCR-centered tasks, facilitating continued progress in this crucial area.
- Level
- Everyday Most adults, without training
- Scored
- Checked automatically A program checks the result against an answer key, unit tests or the expected final state.
- Answers
- Short answer A number, word or short string matched to a key.
2 models share the top score of 83.4% among models released by Apr 2026
Best score per model as labs report them in their own cards. Labs pick their own settings, so these compare less cleanly than independent runs.
The newest model with a score here came out in Apr 2026, and nobody has reported a newer one. Among the 20 highest-ranked models on the capability index, these have no score yet:
- GPT-6 Astra
- Claude Opus 5.5
- GPT-6.1 Terra
- GPT-6.1 Sol
- and 16 more
- 1 Qwen3.5-Omni-Plus 83.4%
- 2 Qwen 3.6 Plus 83.4%
- 3 Qwen3.5 397B-A17B 82%
- 4 Qwen 3.6 35B-A3B 81.9%
- 5 Qwen3.5-122B-A10B 81.8%
- 6 Qwen3-VL-235B-A22B 81.5%
- 7 Qwen3.6 27B 81.2%
- 8 Qwen2.5-VL-72B 79.8%
| # | Model | Score | Setting | Date | Source |
|---|---|---|---|---|---|
| 1 | Qwen3.5-Omni-Plus | 83.4% | Mar 2026 | Model card | |
| 2 | Qwen 3.6 Plus | 83.4% | Apr 2026 | Model card | |
| 3 | Qwen3.5 397B-A17B | 82% | Feb 2026 | Model card | |
| 4 | Qwen 3.6 35B-A3B | 81.9% | Apr 2026 | Model card | |
| 5 | Qwen3.5-122B-A10B | 81.8% | Feb 2026 | Model card | |
| 6 | Qwen3-VL-235B-A22B | 81.5% | Sep 2025 | Model card | |
| 7 | Qwen3.6 27B | 81.2% | Apr 2026 | Model card | |
| 8 | Qwen2.5-VL-72B | 79.8% | Jan 2025 | Model card |
4 models have held number one since its release in Dec 2024
Lab claims count from the day the card came out, independent runs from the model’s release.
Scores over time
All the scores we have, one dot each, so a model tested at several settings appears more than once.
Labs headline it less often now, last in Apr 2026
Launch posts from every lab that has listed it, one tick per post. Tall colored ticks include it, and short gray ones leave it out.
| Date | Launch post | Score | As the post writes it |
|---|---|---|---|
| 22 Apr 2026 | Qwen3.6 27B | 81.2 | CC-OCR |
| 15 Apr 2026 | Qwen 3.6 35B-A3B | 81.9 | CC-OCR |
| 2 Apr 2026 | Qwen 3.6 Plus | 83.4 | CC-OCR |
| 30 Mar 2026 | Qwen3.5-Omni-Plus | 83.4 | CC-OCR |
| 24 Feb 2026 | Qwen3.5-122B-A10B | 81.8 | CC-OCR |
| 16 Feb 2026 | Qwen3.5 397B-A17B | 82 | CC-OCR |
| 23 Sep 2025 | Qwen3-VL-235B-A22B | 81.5 | CC_OCR |
| 28 Jan 2025 | Qwen2.5-VL-72B | 79.8 | CC-OCR |
4 system cards report it, from 2 labs
Besides launch posts, we read the system cards, model cards and technical reports labs publish with a release. They report far more, including safety and frontier-risk tests.
| Date | Document | Score | As the document writes it | Section |
|---|---|---|---|---|
| 17 Apr 2026 | Qwen3.5-Omni-Plus technical report | 83.4 | CC-OCR (Table 6) | Capabilities |
| 14 Jan 2026 |  STEP3-VL-10B technical report | CC-OCR Multi-Lang-OCR track only, SeRe 76.59, PaCoRe 77.51, no overall CC-OCR score (Tables 1 and 3) | Capabilities | |
| 26 Nov 2025 | Qwen3-VL-235B-A22B technical report | 81.5 | CC-OCR (Table 2) | Capabilities |
| 19 Feb 2025 | Qwen2.5-VL-72B technical report | 79.8 | CC-OCR (Table 5) | Capabilities |
Versions
More in Vision
- MMMU In use 11.5K college-level multimodal questions across 30 subjects. Multiple-choice and open accuracy. GPT-5.1 85.4%
- MMMU-Pro In use Harder MMMU: college-level multimodal questions with more options and vision-only inputs to remove text shortcuts. Kimi K3.1 Pro 84.1%
- CharXiv In use Chart understanding on real arXiv figures. Reasoning questions require synthesizing information from complex charts. Claude Fable 5 93.5%
- MATH-Vision (MathVision) In use 3,040 competition math problems with visual context. Answer accuracy. Qwen 3.8 Max 97.7%
- ChartQA In use Models answer human-written and generated questions about chart images that need visual and arithmetic reasoning. Scored by relaxed accuracy with a 5 percent numeric tolerance.
Command A Vision 90.9% - DocVQA In use Models answer 50,000 questions about 12,000 scanned document images by reading their text and layout. Scoring is ANLS, a string similarity to reference answers.
Seed1.5-VL 96.9%