DeepSeek V4 preview drops open weights with a 1M-token context window
DeepSeek published open weights for DeepSeek-V4-Pro (1.6T total, 49B active parameters) and V4-Flash (284B total, 13B active), both with 1M-token context and thinking and non-thinking modes. The API was updated the same day with OpenAI- and Anthropic-compatible endpoints, and older deepseek-chat and deepseek-reasoner names were scheduled for retirement in July. For practitioners, it offered low-cost long-context models and required migrating code to the new model names.
Read the full story at DeepSeek ↗Like, save or discuss on Buzz
