DeepSeek V4 Flash now runs updated weights on AI Gateway

Vercel's AI Gateway now serves DeepSeek V4 Flash with updated weights by default, boosting its Terminal-Bench score to 82.7, up 25.8 points from 56.9 in the April preview. Requests to `deepseek/deepseek-v4-flash` automatically use the new weights with no code changes, and DeepSeek is currently the only provider serving them, with others expected next week.

DeepSeek V4 Flash https://vercel.com/ai-gateway/models/deepseek-v4-flash 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. Requests to deepseek/deepseek-v4-flash pick up the new weights automatically, with no change to the model ID or your code. 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 deepseek/deepseek-v4-flash in the AI SDK https://ai-sdk.dev/ . AI Gateway will route to providers with the new weights by default: To run V4 Flash in a coding agent, use vercel ai-gateway coding-agents setup to connect your agents to AI Gateway, then select deepseek/deepseek-v4-flash in the agent's model configuration. See the coding agents guide https://vercel.com/docs/ai-gateway/coding-agents . AI Gateway reflects provider pricing with no markup and does not charge a platform fee on inference, including on Bring Your Own Key https://vercel.com/docs/ai-gateway/authentication-and-byok/byok BYOK requests. Learn more about AI Gateway https://vercel.com/docs/ai-gateway , view the AI Gateway model leaderboard https://vercel.com/ai-gateway/leaderboards or try it in our model playground https://vercel.com/ai-gateway/models/deepseek-v4-flash .