qwen2.5:14b

Portuguese。

Italian， Japanese， long-text generation (over 8K tokens)， Thai， French。

Vietnamese。

while the 3B and 72B models are under the Qwen license. ， Qwen2.5 is the latest series of Qwen large language models. For Qwen2.5， tables)，。

including Chinese， with sizes ranging from 0.5 to 72 billion parameters. Qwen2.5 introduces the following improvements over Qwen2: It possesses significantly more knowledge and has greatly enhanced capabilities in coding and mathematics 。

a range of base language models and instruction-tuned models are released， understanding structured data (e.g.。

improving role-play and condition-setting for chatbots. It supports long contexts of up to 128K tokens and can generate up to 8K tokens. It offers multilingual support for over 29 languages， Russian， and more. Please note: all models except the 3B and 72B are released under the Apache 2.0 license， English， German， especially in JSON format. It is also more resilient to diverse system prompts ， Korean， due to specialized expert models in these domains. It demonstrates significant advancements in instruction following ， Arabic， Spanish， and generating structured outputs 。