DeepSeek V4.1 Flash 1017 Ultrafast now available on Vercel AI Gateway in public preview
Vercel announced that DeepSeek V4.1 Flash 1017 Ultrafast, the October 17 update of DeepSeek V4.1 Flash served through Fireworks' accelerated serverless endpoint, is available on AI Gateway in public preview. The model has a 1M-token context window and is accessible with no waitroom or session time limit, using the model name deepseek/deepseek-v4.1-flash-1017-ultrafast.
Why it matters
Builders can call a high-throughput DeepSeek model through a single AI Gateway API, with AI Gateway passing through provider pricing with no markup and no platform fee on inference, including BYOK requests. The 1M-token context window suits large codebases and long inputs, and the source describes the model's speed as up to 1,000 tokens per second.
Who should care
Developers building coding tools, agentic workflows, and interactive AI features; teams using Claude Code, Codex, or Cursor with alternative models.
What you can do
Set the model name to deepseek/deepseek-v4.1-flash-1017-ultrafast in an AI Gateway call, or install the latest Vercel CLI and run setup to select it in a coding agent.