DeepSeek models

Access 2 DeepSeek models through the Oxyy unified API including DeepSeek-V4-Pro-0813, DeepSeek-V4.1-Flash. Compare pricing, context windows and capabilities between DeepSeek models.

DeepSeek tokens processed on Oxyy· daily, UTC

Models 2

DeepSeek-V4-Pro-0813TextText-60%107 tokens

reasoningcodinglong-contextflagship

DeepSeek's flagship model, with a 1M-token context window and both thinking and non-thinking modes. Text only.

by DeepSeekAug 13, 20261M context$0.660/M input$1.98/M output3120ms latency16 t/s

DeepSeek-V4.1-FlashTextImageText-60%14d ago

reasoningcodinglong-contextvisionlow-cost+1 more

The smallest model in DeepSeek's new architecture family, with native multimodal visual understanding. Designed for a higher capability ceiling, faster inference, higher throughput, and scaling to larger models.

by DeepSeekSep 10, 20261M context$0.150/M input$0.600/M output