megachangelog
Announcement

DeepSeek V4 Flash now runs updated weights on AI Gateway

DeepSeek V4 Flash has been updated with new weights on Vercel's AI Gateway, delivering significantly stronger agentic capabilities and improved performance scores. The updated model is automatically available to existing requests with no code changes required.

now runs on updated weights by default on AI Gateway, with notably stronger agentic capabilities. On Terminal-Bench, it scores 82.7, up 25.8 points from 56.9 in the April preview.DeepSeek V4 Flash

Requests to pick up the new weights automatically, with no change to the model ID or your code.deepseek/deepseek-v4-flash

For now, DeepSeek is the only provider serving the updated weights. Other providers, including ones with Zero Data Retention, are coming next week.

To use the updated DeepSeek V4 Flash, set model to in the . AI Gateway will route to providers with the new weights by default:deepseek/deepseek-v4-flashAI SDK

To run V4 Flash in a coding agent, use to connect your agents to AI Gateway, then select in the agent's model configuration. See the .vercel ai-gateway coding-agents setupdeepseek/deepseek-v4-flashcoding agents guide

AI Gateway reflects provider pricing with no markup and does not charge a platform fee on inference, including on (BYOK) requests. Learn more about , view the or try it in our .Bring Your Own KeyAI GatewayAI Gateway model leaderboardmodel playground

Read more

ai-gatewaydeepseekmodelsperformance

Source: original entry ↗