Unverified50% confidenceFactExact time
E2AM会标明每个数值是实测还是估算，若GPU未暴露NVML功耗传感器则退回到功率上限×利用率的估算方式
1
Sources
50%
Confidence
Medium-term (~90 days)
Relevance
7/12/2026
First Seen
Valid until: 10/10/2026
Sources
E2AM：两行代码量化AI训练能耗与碳排放的开源工具
redditr/deeplearning7/11/2026
Related Claims
UnverifiedGPTQ、AWQ等后训练量化方法能够在几乎不损失基准测试分数的情况下将70B参数量级模型部署在消费级GPU上63% similarUnverified为关键任务建立对AI输出质量的持续监控并保留不同版本模型的对比能力有助于及时发现潜在退化问题62% similarUnverifiedQAOA在当前硬件条件下相对经典优化方法的实质优势尚需更系统的基准测试支撑，R²结果也有待在独立数据集上进一步验证62% similarUnverified系统识别通过测量真实硬件参数来校准仿真模型61% similarUnverifiedThe SM2 algorithm maintains an 'easiness factor' (EF value) for each learning item and dynamically adjusts review intervals based on recall quality scores.60% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/491604API
curl https://kongchang.com/api/v1/knowledge/claims/491604MCP
get_claim(id=491604)