DeepSeek V4 Pro 0813 launches with 1M context and $0.435 per million input tokens

DeepSeek's V4 Pro 0813, the GA release of V4 Pro, is now available on OpenRouter. This large-scale mixture-of-experts model supports a 1M-token context window and is priced at $0.435 per million input tokens and $0.87 per million output tokens. It features hybrid attention for efficient long-context processing and supports reasoning efforts up to 'xhigh', making it suitable for complex workloads like full-codebase analysis and multi-step automation. The model is hosted solely by DeepSeek, with caching bringing the average effective input price down to $0.1537 per million tokens.