Pangu 3.0

Bylo to zveřejněno v July 2023. Schopnost na parametr se od té doby výrazně zlepšila, takže novější model stejné velikosti je často lepším využitím stejného hardwaru.

funguje v oblasti Multimodal, Language, Image generation, Vision, a je zaznamenáno jako zpracovávající úkol language modeling/generation, Image generation. Model může nést několik z každého, takže to jsou oblasti, pro které byl navržen, spíše než omezení toho, co se pokusí.

Žádné. Toto je uzavřený model — jeho váhy nikdy nebyly zveřejněny, takže jej nelze stáhnout nebo spustit na vlastním hardwaru za žádnou cenu. Je dostupný pouze prostřednictvím jeho poskytovatele.

Ne. Jeho váhy nebyly zveřejněny, takže existuje pouze jako služba řízená jeho vlastníkem.

Má počet parametrů 100B. 100B? I think the five foundation models are all included in the same system, instead of being five different variants of Pangu, but that's not very clear. I think that's implied by "All of these capability sets will be made available to customers and partners, and will be consistent regardless of the model size". https://www.huaweicloud.com/intl/en-us/news/20230707180809498.html "Huawei Cloud Pangu models were designed to focus on the practical needs of specific industry scenarios. The newly-launched Pangu Models 3.0 use a "5+N+X" three-layer architecture. The L0 layer consists of five foundation models: NLP, CV, multimodal, prediction, and scientific computing, which provide general skills to power an endless possibility of industry-specific applications. Pangu Models 3.0 are available in different sizes: 10 billion parameters, 38 billion parameters, 71 billion parameters, and 100 billion parameters, meeting diverse customer needs and different standards on latency and response times. Brand-new capability sets are also provided, such as knowledge-based Q&A, copywriting, and code generation for the Pangu NLP model; and image generation and understanding for the Pangu multimodal model. All of these capability sets will be made available to customers and partners, and will be consistent regardless of the model size (number of parameters). The L1 layer consists of N industry-tailored models. Huawei Cloud can provide customers with industry models it has trained on open industry datasets, including Pangu models for government, finance, manufacturing, mining, and meteorology. Alternatively, customers can train their own models using their own datasets based on Huawei's L0 or L1 Pangu models. The L2 layer provides pre-trained models for specific industry scenarios and tasks, such as intelligent government hotline, intelligent branch assistant, lead compound screening, conveyor belt foreign object detection, and typhoon trajectory prediction. These models can be quickly deployed off-the-shelf.". Tato cifra je celková a rozhoduje o tom, kolik VRAM model potřebuje — přibližně půl gigabajtu na miliardu při kompresi, kterou většina lidí používá.

Bylo to publikováno společností Huawei, zakládající se na China, organizace zařazená jako industry.