Vercel Adds GLM-5.3 Flash to AI Gateway, Here’s What It Supports
Quick Reads
- Vercel has added GLM-5.3 Flash from Z.ai to its AI Gateway platform.
- The model supports text, vision input, function calling, structured output, and streaming.
- It comes with a 1 million token context window.
- It works with coding agents like Claude Code, Codex, Cursor, and more.
- Vercel charges no platform fee on inference, including on Bring Your Own Key requests.
Vercel has launched support for GLM-5.3 Flash from Z.ai on its AI Gateway. The addition brings a capable multimodal model to developers building on Vercel’s infrastructure.
GLM-5.3 Flash is a multimodal model that handles both text and vision input. It ships with a 1 million token context window, which is notably large. Beyond that, it supports function calling, structured output, and streaming, covering most use cases modern developers care about.
To get started, developers simply set the model parameter to zai/glm-5.3-flash in their project. The model also accepts images alongside text. Both URLs and Base64 data URLs work, and a single request can include multiple images at once.
GLM-5.3 Flash also integrates with Vercel’s coding agents. Developers can run vercel ai-gateway coding-agents setup to connect it with agents such as Claude Code, Codex, OpenCode, Cursor, Pi, and others. From there, they select zai/glm-5.3-flash within the agent to activate it.
Vercel AI Gateway, as a platform, provides a unified API for calling models and tracking usage and cost. It also supports retries, failover, and performance optimizations designed for uptime above the provider level. The platform includes custom reporting, Zero Data Retention support, API key budgets, and routing rules.
On pricing, Vercel reflects provider pricing without any markup. There is no platform fee on inference, including on Bring Your Own Key (BYOK) requests. For Nigerian and African developers building AI-powered applications, this cost structure lowers the barrier to accessing frontier models through a professional-grade API layer.
GLM-5.3 Flash joins a growing list of language models available on AI Gateway. Developers can test it directly in the model playground.
