This week Vercel added one of the fastest reasoning models to AI Gateway. DeepSeek V4.1 Flash 1017 Ultrafast clocks over 1,000 tokens per second and brings a 1M-token context window, built for coding, reasoning and agentic workflows.
DeepSeek V4.1 Flash 1017 Ultrafast
The model is available in public preview with no waitroom or session limits, through the Vercel AI Gateway and Fireworks. DeepSeek V4.1 Flash 1017 Ultrafast is now on AI Gateway.
Call it with the model name deepseek/deepseek-v4.1-flash-1017-ultrafast in the AI SDK, or set it up via the Vercel CLI to route it to coding agents like Claude Code and Cursor.
- Speed
- 1,000+ tokens per second
- Context window
- 1M tokens
Sources
2 checkedHow we cover tool news: Create With's tool desk drafts these reports with AI from the sources listed above and checks them against those sources before publishing.







