- 1,000+
- Tokens per second
- 1M
- Context window (tokens)
- Price
- No platform fee
- Who it's for
- Developers
- When
- Available now
- Where it works
- Vercel AI Gateway
Key points
- Upwards of 1,000 tokens per second
- Built for interactive coding and agents
- No markup or fee on inference
- Public preview with no waitlist
Where to use it
- Software developmentUse it in coding agents such as Cursor
- Automated workflowsChain steps that need quick model replies
- Long documentsFeed in large files that fit the 1M-token window
What it does
Gives developers a DeepSeek model that generates upwards of 1,000 tokens per second, through AI Gateway from Vercel.
Who it's for
Developers building interactive coding tools, reasoning steps and agents that need quick replies.
Price and where to try it
Vercel says AI Gateway adds no markup and no platform fee on inference. It is in public preview with no waitroom, and you can try it in the model playground.
Written with AI from the source above. If something looks off, the source is what counts.
