DeepSeek V4 is now available on Vercel's AI Gateway with two model variants: DeepSeek V4 Pro and DeepSeek V4 Flash. Both offer a 1M token context window. V4 Pro targets agentic coding, mathematical reasoning, and long-horizon workflows with MCP and agent framework support. V4 Flash offers faster responses and lower cost for high-volume, latency-sensitive workloads. Access is available via the AI SDK using model identifiers `deepseek/deepseek-v4-pro` or `deepseek/deepseek-v4-flash`, with no markup and no separate provider account required.
Sort: