Found a total of 4 related content
Yi-VL large model is open source and ranks first in MMMU and CMMMU
Article Introduction:On January 22, the Yi series model family welcomed a new member: YiVisionLanguage (Yi-VL) multi-modal language large model was officially open sourced globally. It is reported that the Yi-VL model is developed based on the Yi language model and includes two versions, Yi-VL-34B and Yi-VL-6B. Yi-VL model open source address: https://huggingface.co/01-aihttps://www.modelscope.cn/organization/01ai With its excellent image and text understanding and dialogue generation capabilities, the Yi-VL model has achieved great results in the English dataset MMMU and achieved leading results on the Chinese data set CMMMU, showing
2024-01-22comment109
零一万物Yi-1.5来了,国产开源大模型排行榜再次刷新
Article Introduction:在OpenAI发布会的前一天,来自中国的大模型公司零一万物,让开源大模型社区活跃起来:Yi大模型家族新成员,“Yi-1.5”模型正式开源。这一举措将进一步推动开源大模型社区的发展,为人工智能技术的普及和应用创造更多机会。在项目主页可以看到,Yi-1.5包括一系列预训练和微调模型,分为6B、9B、34B三个版本,采用Apache2.0许可证。GitHub地址:https://github.com/01-ai/Yi-1.5HuggingFace模型下载地址:https://huggingface.c
2024-05-14comment 0285
Kai-fu Lee's AI company 'Zero One Thousand Things' open source Yi large model was accused of plagiarism LLaMA
Article Introduction:Kai-fu Lee, chairman and CEO of Sinovation Ventures, founded the AI large-model startup "Zero One World" this year. The company has released two large open source models, Yi-34B and Yi-6B, and claimed that it is completely open to academic research and allows free commercial applications. On HuggingFace’s open source homepage, ehartford, the developer of the Yi-34B model, raised questions, saying The model uses the MetaLLaMA architecture, and only the names of two tensors (Tensor) have been modified, namely input_layernorm and post_attention_layernormehartford said that the open source community will definitely re-release Y
2023-11-14comment 0578
Consumer grade graphics cards available! Li Kaifu released and open sourced the 9 billion parameter Yi model, which has the strongest code mathematical ability in history
Article Introduction:Kai-Fu Lee’s AI company, Zero One, has another big model player on the scene: the 9 billion parameter Yi-9B. It is known as the "science champion" in the Yi series, "advancing" code mathematics, and at the same time, its comprehensive ability has not fallen behind. The best performance among a series of open source models of similar scale (including Mistral-7B, SOLAR-10.7B, Gemma-7B, DeepSeek-Coder-7B-Base-v1.5, etc.). The old rule is that it is open source when released, which is especially friendly to developers: Yi-9B (BF16) and its quantitative version Yi-9B (Int8) can be deployed on consumer-grade graphics cards. An RTX4090 or an RTX3090 is enough. Deep augmentation + multi-stage incremental training
2024-03-07comment 0502