The team behind Vercel’s V0, an AI-driven platform for web creation, has developed an AI model that claims to be superior to specific website development tasks.
The model called “V0-1.0-MD”, available via the API, can be prompted with text or images, and “was optimized for front-end and full-stack web development,” says the Vercel team. Currently, the beta requires a V0 Premium Plan ($20 per month) or a Team Plan ($30 per month).
The launch of the V0 model is because more developers and businesses are looking to adopt AI-powered tools for programming. According to a Stack Overflow survey last year, around 82% of developers reported using AI tools to write code. Meanwhile, a quarter of Y-combinator’s W25 batch startups have 95% of the AI-generated codebase, according to YC managing partner Jared Friedman.
Vercel’s model can “auto-fix” common coding problems, and according to the Vercel team, it is compatible with tools and SDKs that support Openai’s API format. When evaluated in a web development framework like next.js, the model can take up to 128,000 tokens at a time.
A token is a bit of data on which the AI model operates, with one million tokens equivalent to about 750,000 words (approximately 163,000 words more than “war and peace.”
Versel is not the only outfit that develops tailored models for programming, and should be careful. Last month, JetBrains, the company behind the range of popular app development tools, debuted its first “open” AI coding model. Last week, Windsurf released a family of programming-focused models called SWE-1. And yesterday, Mistral announced Devstral, a model tailored to specific developer tasks.
While companies may be keen to develop and accept AI-powered coding assistants, models still struggle to create high-quality software. Code generation AI tends to introduce security vulnerabilities and errors due to weaknesses in areas such as the ability to understand programming logic.