Hello Qwen2
Qwen
Alibaba released Qwen2, the successor to Qwen1.5, with five model sizes ranging from 0.5B to 72B parameters. The models support 29 languages total, achieve state-of-the-art results on multiple benchmarks, and the largest versions handle context windows up to 128K tokens. The release expands Alibaba's language model offerings with improved capabilities in coding, mathematics, and multilingual support.
Why it matters
GITHUB HUGGING FACE MODELSCOPE DEMO DISCORD Introduction After months of efforts, we are pleased to announce the evolution from Qwen1.5 to Qwen2. This time, we bring to you: Pretrained and instruction-tuned models of 5 sizes, including Qwen2-0.5B, Qwen2-1.5B, Qwen2-7B, Qwen2-57B-A14B, and Qwen2-72B; Having been trained on data in 27 additional languages besides English and Chinese; State-of-the-art performance in a large number of benchmark evaluations; Significantly improved performance in coding and mathematics; Extended context length support up to 128K tokens with Qwen2-7B-Instruct and Qwen2-72B-Instruct.