DeepSeek V4 Pro now runs updated weights on AI Gateway
DeepSeek V4 Pro gets a quiet weight update on Vercel's AI Gateway — agents and coding workflows pick it up by default.
Why it matters
Vercel's AI Gateway rolls out updated DeepSeek V4 Pro weights with seamless routing for existing code. Practitioners using DeepSeek models via Gateway get the new version automatically; coding agents can opt in explicitly. No platform markup or inference fees.
The key facts
11 to knowDeepSeek V4 Pro updated weights now live on Vercel AI Gateway
Existing requests route to new weights by default (no code change required)
New model ID available: deepseek/deepseek-v4-pro-0813 for explicit pinning
Coding agents can connect via ai-gateway and select updated model in configuration
AI Gateway pricing: provider pricing passed through with no platform markup or inference fees
BYOK (Bring Your Own Key) requests supported with no additional platform fee
DeepSeek V4 Pro updated weights (0813 version) now default on Vercel AI Gateway
Existing code picks up new weights automatically with no model ID change
New coding-agent integration path: use vercel ai-gateway + select deepseek-v4-pro-0813 in agent config
AI Gateway reflects provider pricing with no platform markup or inference fee
Model leaderboard and playground available for testing
Go to the source
Vercel Blogvercel.com
Publisher excerpt: DeepSeek V4 Pro now runs on updated weights on AI Gateway. They are used by default when you call , so existing requests pick them up with no change to the model ID or your code.deepseek/deepseek-v4-pro To use the updated DeepSeek V4 Pro, set model to in the . AI Gateway will route to providers…