The code generation capability has not reached its goal. Alphabet Inc. Class C (GOOGL.US) delays the release of the Gemini 3.5 Pro large language model.

Due to the delayed announcement of the flagship artificial intelligence model, Alphabet (GOOGL.US, GOOG.US) stock fell by 4% on Thursday. According to media reports, Alphabet Inc. Class C has decided to postpone the release of Gemini 3.5 Pro, a large language model, for several months as the model's performance has not met internal expectations. Sources revealed that the development progress of Gemini 3.5 Pro has fallen behind schedule, mainly because Alphabet Inc. Class C is still optimizing the model's performance. The capability of code generation has not reached the company's internal goals, and this area has become one of the most fiercely competitive races among global AI companies. As Alphabet Inc. Class C slows down its product release schedule, OpenAI and Meta Platforms (META.US) have recently introduced new generation AI models with better code generation capabilities than Alphabet Inc. Class C's existing products, further intensifying market competition. In May of this year, Alphabet Inc. Class C first introduced Gemini 3.5 Pro at the Google I/O developer conference and stated that the model had already started being used internally at the company, with plans to open it up to more users in the following month. However, the latest news shows that this plan has been delayed. In response to the reports, an Alphabet spokesperson stated that the company continues to rapidly launch multiple AI models while maintaining cost efficiency. The spokesperson mentioned that the company is currently testing Gemini 3.5 Pro, the upgraded version of the Flash model, and other AI models with partners, and is actively collaborating with the U.S. government. In recent years, code generation has become one of the most important applications of generative AI and a key focus of competition among major AI model providers. Companies such as OpenAI and Anthropic have been continuously releasing code generation models for developers, some of which are open source and available for free. Just last week, Meta released the new generation large language model Muse Spark 1.1. Meta's Chief AI Officer, Alexandr Wang, stated that this is the company's strongest model in the AI agent and code generation field to date. Similarly, OpenAI launched the GPT-5.6 Sol model last week. OpenAI's CEO, Sam Altman, stated that the new model has improved efficiency in token usage by 54% in AI agent code generation tasks, further strengthening the model's competitive advantage in performance and cost. Analysts believe that as AI model competition shifts from parameter scale to actual application capabilities and cost efficiency, code generation capability has become an important indicator of competitiveness for large language models. The delayed release of Gemini 3.5 Pro reflects Alphabet Inc. Class C's desire to officially launch the model after its performance matures in order to cope with the increasingly intense industry competition.