Beijing Zero One Everything Technology Co., Ltd. officially announces the open-source release of the Yi-Coder series model, marking another significant advancement for the company in the field of artificial intelligence. The Yi-Coder series focuses on programming tasks, offering models with 1.5B and 9B parameter scales, with Yi-Coder-9B demonstrating exceptional performance in code generation, understanding, debugging, and completion.

Recently, Yi API under Zero One Everything 01AI announced the launch of the Yi-Large-FC model with Function Call capability. This new feature enables the model to determine when to call external tools based on user input and respond in JSON format, compatible with OpenAI's interface design, achieving a smooth alternative to GPT's capabilities.

On June 25th, OpenAI's announcement of supply cuts caused considerable disruption to developers and businesses. In response to this move by OpenAI, the AI large model company Zero One Everything, founded by Dr. Kai-Fu Lee, quickly reacted by launching the 'Yi API 80% Discount Alternative Plan.' This service is aimed at OpenAI users, designed to help them seamlessly transition to the Yi series large models, providing a cost-effective alternative.

Zero One Everything has announced the open sourcing of the Yi-9B model. Yi-9B is the strongest model in the Yi series in terms of coding and mathematical capabilities. It excels in overall, coding, and mathematical abilities, and can be easily deployed on consumer-grade graphics cards. The company was founded by Kaifu Lee, Chairman and CEO of Innovation Works.

The Yi-VL multimodal language model has been launched, including two versions: Yi-VL-34B and Yi-VL-6B. The Yi-VL model demonstrates exceptional capabilities in visual-text understanding and dialogue generation. It has achieved leading performance on English and Chinese datasets. Yi-VL-34B surpasses other multimodal large models with an accuracy rate of 41.6%. The Yi-VL model is based on the LLaVA architecture, featuring strong language understanding and generation abilities.

The large model Yi-34B uses the LLaMA architecture but has changed tensor names, leading to controversy. Model performance includes a context window size exceeding 200k, processing 400,000 Chinese characters at once, and achievements in the Hugging Face Global Open Source Model Rankings. This has led the community to question whether Yi-34B has substantively altered the code and whether there is false advertising and license violations. The Zero One Everything company responded that the changes were based on the GPT structure, and that training requirements led to the renaming of code. The controversy focuses on whether the model violates open source agreements and the transparency of performance parameters and code modifications.