Retour aux modèles
Alibaba Qwen
Qwen2.5
Open source3 variantes
Alibaba's 14B Qwen 2.5 instruct model — strong coding, math, and instruction-following at a mid-size, cost-efficient deployment point.
131K tokensGratuit / Poids ouvertsTransformerApache 2.0
Qwen2.5-14B-Instruct
42.6%Qwen2.5-14B-Instruct FC — estimated from Qwen2.5 family (between 7B and 32B scores)
instruct72B
Alibaba's 72B Qwen 2.5 flagship instruct model — top-tier open-source performance on coding, math, and instruction-following.
131K tokensGratuit / Poids ouvertsTransformer
instruct7B
Alibaba's 7B Qwen 2.5 instruct model — top-tier small model for chat, coding, and math with a 128K context window.
131K tokensGratuit / Poids ouvertsTransformer
