Chinese AI labs take most top spots in a web coding ranking
Arena’s web development code compares AI models on front-end web development tasks and multi-step automated coding work. As of July 1, 2026, it was based on 422,204 votes across models from 18 labs.
By lab ranking, the top five are Anthropic’s claude-fable-5 in first place, Z.ai’s glm-5.2 in second, ByteDance’s seed-2.1-pro-preview in third, Alibaba’s qwen3.7-max-20260517 in fourth, and Moonshot’s kimi-k2.6 in fifth. Anthropic is the only non-Chinese lab in that top five group.
Google’s gemini-3.5-flash is sixth, OpenAI’s gpt-5.5-xhigh is eighth, and DeepSeek’s -pro-thinking is tenth. Some models are marked as , and the ranking is focused on web development, so it should not be treated as a full measure of every coding use case.
Key points
- Arena ranks AI models for web development using vote-based comparisons.
- The July 1, 2026 ranking includes 422,204 votes and 18 labs.
- Four of the top five labs are Chinese: Z.ai, ByteDance, Alibaba, and Moonshot.
- Google ranks sixth, OpenAI ranks eighth, and DeepSeek ranks tenth on this page.
- The result is specific to web development tasks, not all programming work.