DeepSeek V4 và mô hình mới của Tencent Huan Yuan dự kiến ra mắt vào tháng Tư

The White Whale Lab reported that both DeepSeek V4 and the new Hunyuan model led by Yao Shunyu are expected to be released in April. According to the report, DeepSeek-V4, a multimodal large model long refined by Liang Wenfeng, will not only see a significant leap in coding capabilities but also achieve breakthroughs in LTM (long-term memory). Recently, OpenRouter launched two mysterious models: Healer Alpha and Hunter Alpha, speculated to be DeepSeek V4. Users with domestic IPs can directly access these two large models. According to their self-introductions, Hunter Alpha is a trillion-parameter model with a 1 million token context window, specifically designed for agent workflows, long-cycle complex tasks, and deep tool invocation. Healer Alpha is a full-modal interactive model with a 260,000-token context window, capable of natively perceiving visual and audio inputs, performing cross-modal reasoning, and executing complex multi-step tasks accurately and reliably. As of publication, neither of these two large models has been officially claimed. Additionally, The White Whale Lab reported that Yao Shunyu, Tencent’s Chief AI Scientist, will also release a new Hunyuan model (30B parameter class) in April. This will be his first major product since joining Tencent in December last year and is highly anticipated.