Qwen2.5-Turbo
Alibaba · 2024-11-15
Qwen2.5-Turbo is Alibaba's dated API-served text-generation snapshot, released on November 15, 2024 as qwen-turbo-2024-11-01. It supports a 1,000,000-token context window, and Qwen reports that sparse attention was used to accelerate inference on 1,000,000-token sequences while maintaining the Turbo model's existing API price.