Unverified50% confidenceFactExact time
据作者截至2026年10月的观察，Qwen上代有0.6B到9B的小模型适合手机和轻量PC，27B到32B在24GB显存的GPU上舒适运行。
1
Sources
50%
Confidence
Medium-term (~90 days)
Relevance
10/12/2026
First Seen
Valid until: 1/10/2027
Sources
本地大模型选型指南：从Fate视角看懂LLM部署
Related Entities
Cite This Claim
Stable URI
https://kongchang.com/claim/1007465API
curl https://kongchang.com/api/v1/knowledge/claims/1007465MCP
get_claim(id=1007465)