Alibaba Cloud has announced a price reduction for the implicit caching of its DeepSeek-V4-Pro model on the Bailian platform. Effective April 29, 2026, at 23:59:59 Beijing Time, the cost will be reduced to RMB 1 per million tokens. This pricing adjustment applies only to requests that hit the cache, with matched input tokens charged as cached_tokens. Unmatched tokens will continue to incur the standard input_token rate. The base inference pricing for the model remains unchanged.