DeepSeek Postpones V4 Model Launch Amid US Scrutiny
- DeepSeek has delayed the launch of its V4 model due to integration with Huawei Ascend chips.
- The V4 model is designed for cost-efficiency, priced at $1.74 per million input tokens and $3.48 per million output tokens.
- Despite being cheaper, the V4 model is expected to lag behind competitors by three to six months.
- US authorities are investigating DeepSeek over allegations related to prohibited technologies and potential national security risks.
- Negotiations with Tencent and Alibaba are ongoing for potential investments in DeepSeek’s technology.
DeepSeek’s decision to focus on local chip integration reflects China’s strategy for technological independence amidst US scrutiny over its operations. The startup’s competitive pricing and innovative architecture aim to enhance its position in the AI market.
As DeepSeek faces challenges from US investigations, it continues to develop its V4 model, which is projected to be three to six months behind leading models in the industry.(Source)
