DeepSeek V4.1 Flash 1017 now available on AI Gateway

DeepSeek V4.1 Flash 1017 is now available on AI Gateway as a minor update to the previous DeepSeek V4.1 Flash snapshot. This release focuses on enhancing performance in reasoning, coding, agentic tasks, and image understanding.

Performance Improvements

The update delivers specific gains in practical development workflows:

• Coding: Improvements are visible in terminal work and when building or editing entire repositories.
• Agentic Tasks: The model shows better performance in multi-step automation.
• Image Understanding: It reads charts and handles visual reasoning more accurately, particularly when calling tools.

Additionally, the model finishes tasks faster and generates shorter outputs compared to the previous version. This results in agent runs that take less time and consume fewer output tokens.

Access and Integration

Users can access the model using the identifier deepseek/deepseek-v4.1-flash-1017. It is compatible with tools like Claude Code, Codex, and Cursor via the latest Vercel CLI setup. A model playground is also available for testing.

AI Gateway Features

AI Gateway provides a unified API for model calls, usage tracking, and cost management. Key features include:

• Built-in custom reporting and Zero Data Retention support.
• Budgets for API keys and routing rules.
• Configurable retries and failover for higher uptime.
• No markup on provider pricing and no platform fee on inference, including Bring Your Own Key (BYOK) requests.