# Gemini 4 Argon now available on Vercel AI Gateway

published: 2026-10-13T17:06:21.642Z
topic: infrastructure
breaking: true
source: https://vercel.com/changelog/gemini-4-argon-now-available-on-ai-gateway
url: https://aiagentnews.io/stories/gemini-4-argon-now-available-on-vercel-ai-gateway

Google's Gemini 4 Argon is now available through Vercel's AI Gateway under the model name google/gemini-4-argon. The model is aimed at coding, agentic work and complex reasoning, and it uses thinking levels in place of thinking budgets.

Vercel says Gemini 4 Argon from Google can be reached through AI Gateway. The model has a 1M token context window, accepts text, image, PDF and video input, and supports tool calling. Vercel positions it for coding, agentic work and complex reasoning, including multi-step coding tasks and analysis of dense spreadsheets and documents.

Developers can call the model through the AI SDK, the Chat Completions, Responses and Anthropic Messages APIs, or a coding agent connected to the gateway. Instead of thinking budgets, the model takes low, medium or high thinking levels, set with reasoning in the AI SDK or reasoning_effort in Chat Completions. Coding agents such as Claude Code, Codex and Cursor can use it after the latest Vercel CLI is installed and the AI Gateway setup command is run.

AI Gateway provides a unified API for models, with usage and cost tracking, retries, failover, custom reporting, Zero Data Retention support, budgets for API keys and routing rules. Vercel says it passes through provider pricing with no markup and charges no platform fee on inference, including Bring Your Own Key requests.

## Why it matters if you build agents

Agent developers can try Gemini 4 Argon through a gateway they may already use, switching models by changing the model name and thinking level setting. Because Vercel adds no platform fee on inference, the cost comes down to provider pricing, which is worth checking against any API key budgets before routing production traffic to the model.