DeepSeek V4-Pro AI Model Launched: Features & Pricing

DeepSeek has officially launched its V4-Pro AI model, now available on its app, web interface, and API. This follows a preview period that began in April. Here's the thing: The V4-Pro model, designated DeepSeek-V4-Pro-0813, focuses on agent capabilities. This means it can use tools, execute code, and complete multi-step tasks without human help. Benchmark results show it scored 87.9 on Terminal Bench 2.1 and 62.7 on DeepSWE. It can handle a context window up to 1 million tokens and produce outputs as long as 384,000 tokens. What's interesting is a price increase for the V4 model family will take effect on August 16th. DeepSeek is also introducing peak and off-peak billing, with off-peak rates at half the peak-hour price. For example, V4-Pro output tokens will rise to $3.96 per million at peak hours, up from the current flat rate of $0.87 per million. The bottom line: DeepSeek is adjusting its pricing and expanding features, which could impact how developers and businesses use its AI models.