Unverified80% confidenceFactTime unknown
在三台手机离线测试中，iQOO 15运行Gemma 4 1/2B模型的速度最快
1
Sources
80%
Confidence
Medium-term (~90 days)
Relevance
6/3/2026
First Seen
Valid until: 9/1/2026
Sources
Google Gemma 4实测：手机离线运行+Ollama部署教程
bilibiliAGI_Ananas
Related Entities
Related Claims
Unverified在本地推理速度对比中，Agent A1（95 tokens/秒）快于Gemma 4 Coder和Qwen系列64% similarUnverifiedMiniCPM 系列以 2B-4B 级别的参数量在多项基准测试中实现了接近 7B 甚至 13B 模型的性能61% similarUnverified树莓派5配合量化模型可运行3B参数级别语言模型，NVIDIA Jetson Orin系列最高275 TOPS AI算力可运行7B甚至13B量化模型，延迟可控制在1秒以内58% similarUnverifiedClaude Haiku 4.5在三个前端测试场景中生成速度最快，成本最低58% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/38373API
curl https://kongchang.com/api/v1/knowledge/claims/38373MCP
get_claim(id=38373)