CVTG-2K

CVTG-2K (Chinese Visual Text Generation 2K) is a benchmark for evaluating text-to-image models on their ability to accurately render text within generated images. It measures Word Accuracy, Normalized Edit Distance (NED), and CLIPScore across 2,000 prompts.

GLM-Image from Zhipu AI currently leads the CVTG-2K leaderboard with a score of 0.912 across 1 evaluated AI models.

About this benchmark

What CVTG-2K measures

CVTG-2K is a image benchmark that evaluates large language models on image-generation, language, and vision tasks. LLM Stats tracks 1 model on this benchmark, with a maximum possible score of 1. Current average across reported models is 0.9, with the leader reaching 0.9.

Compare leaders on the best AI for image-generation, best AI for language and best AI for vision leaderboards.

Zhipu AIGLM-Image leads with 91.2%.

Progress Over Time

Interactive timeline showing model performance evolution on CVTG-2K

State-of-the-art frontier
Open
Proprietary

CVTG-2K Leaderboard

1 models
ContextCostLicense
1
Zhipu AI
Zhipu AI
16B
Notice missing or incorrect data?

FAQ

Common questions about CVTG-2K.

What is the CVTG-2K benchmark?

CVTG-2K (Chinese Visual Text Generation 2K) is a benchmark for evaluating text-to-image models on their ability to accurately render text within generated images. It measures Word Accuracy, Normalized Edit Distance (NED), and CLIPScore across 2,000 prompts.

What is the CVTG-2K leaderboard?

The CVTG-2K leaderboard ranks 1 AI models based on their performance on this benchmark. Currently, GLM-Image by Zhipu AI leads with a score of 0.912. The average score across all models is 0.912.

What is the highest CVTG-2K score?

The highest CVTG-2K score is 0.912, achieved by GLM-Image from Zhipu AI.

How many models are evaluated on CVTG-2K?

1 models have been evaluated on the CVTG-2K benchmark, with 0 verified results and 1 self-reported results.

What categories does CVTG-2K cover?

CVTG-2K is categorized under image-generation, language, and vision. The benchmark evaluates image models.

What is the best open-source model on CVTG-2K?

GLM-Image by Zhipu AI is the top-ranked open-source model on CVTG-2K, with a score of 0.912 (rank #1).

How recent are the CVTG-2K leaderboard results?

The CVTG-2K leaderboard was last updated in June 2026 and currently includes 1 evaluated models.

More evaluations to explore

Related benchmarks in the same category

View all image-generation
MMLU-Pro

A more robust and challenging multi-task language understanding benchmark that extends MMLU by expanding multiple-choice options from 4 to 10, eliminating trivial questions, and focusing on reasoning-intensive tasks. Features over 12,000 curated questions across 14 domains and causes a 16-33% accuracy drop compared to original MMLU.

language
127 models
MMLU

Massive Multitask Language Understanding benchmark testing knowledge across 57 diverse subjects including STEM, humanities, social sciences, and professional domains

language
100 models
Humanity's Last Exam

Humanity's Last Exam (HLE) is a multi-modal academic benchmark with 2,500 questions across mathematics, humanities, and natural sciences, designed to test LLM capabilities at the frontier of human knowledge with unambiguous, verifiable solutions

visionmultimodal
82 models
MMMU

MMMU (Massive Multi-discipline Multimodal Understanding) is a benchmark designed to evaluate multimodal models on college-level subject knowledge and deliberate reasoning. Contains 11.5K meticulously collected multimodal questions from college exams, quizzes, and textbooks, covering six core disciplines: Art & Design, Business, Science, Health & Medicine, Humanities & Social Science, and Tech & Engineering across 30 subjects and 183 subfields.

visionmultimodal
62 models
MMMU-Pro

A more robust multi-discipline multimodal understanding benchmark that enhances MMMU through a three-step process: filtering text-only answerable questions, augmenting candidate options, and introducing vision-only input settings. Achieves significantly lower model performance (16.8-26.9%) compared to original MMMU, providing more rigorous evaluation that closely mimics real-world scenarios.

visionmultimodal
56 models
MMMLU

Multilingual Massive Multitask Language Understanding dataset released by OpenAI, featuring professionally translated MMLU test questions across 14 languages including Arabic, Bengali, German, Spanish, French, Hindi, Indonesian, Italian, Japanese, Korean, Portuguese, Swahili, Yoruba, and Chinese. Contains approximately 15,908 multiple-choice questions per language covering 57 subjects.

language
48 models