OpenAI首席执行官奥特曼过去曾表示，整体计算能力也是最新一波人工智能模型的一个问题。奥特曼最近在Reddit上表示：“所有这些模式都变得相当复杂，我们无法同时推出那么多东西。”“我们还面临着许多限制和艰难的决定，即如何将我们的计算分配给许多伟大的想法。”Like OpenAI, Anthropic, supported by google (GOOGL.US) and amazon (AMZN.US), also stated that the performance of its large language models has stalled.
According to informed sources reported by Zhitongcaijing APP, like OpenAI, Anthropic, supported by google (GOOGL.US) and amazon (AMZN.US), also stated that the performance of its large language models has stalled.
Three informed sources said that the upcoming Gemini version failed to meet internal expectations. In addition, Anthropic has also postponed the next version of its Claude model, which is 3.5 Opus. One of the issues these companies face is the lack of human data for training models.
OpenAI CEO Altman has previously stated that overall computing power is also an issue for the latest wave of ai models. Recently, Altman said on Reddit: "All these patterns have become quite complex, and we can't launch that many things at once. We also face many constraints and tough decisions about how to allocate our computing resources to many great ideas."
Last month, Anthropic released a new version called Claude 3.5 Haiku, along with an upgraded version of Claude 3.5 Sonnet.
According to informed sources, OpenAI, supported by microsoft (MSFT.US), encountered performance issues with its upcoming large language model, code-named Orion. OpenAI has previously denied that it would release a model named Orion.
In September of this year, OpenAI, led by Altman, launched a series of new ai models, named o1 and o1-mini, which can reason through complex tasks and solve more difficult problems than previous science, coding, and math models.
