Verified65% confidenceFactExact time
The energy consumption during the training phase of large models is approximately hundreds of times that of the inference phase.
3
Sources
65%
Confidence
Long-term
Relevance
7/12/2026
First Seen
Sources
AI Data Centers Competing for Power: Why U.S. Manufacturing Energy Costs Keep Soaring
hackernewshackernews7/7/2026
Related Claims
UnverifiedReasoning models are extremely costly to train, typically requiring a combination of large-scale reinforcement learning and carefully designed reward models to acquire the capability.80% similarVerifiedTraining frontier large models requires enormous computational investment, and there are diminishing marginal returns as model scale increases.78% similarUnverifiedExact unlearning achieves complete deletion by retraining the model after removing the target data, which is extremely costly for large models with billions of parameters, as a single complete training run may consume millions of dollars in computing resources.76% similarUnverifiedIn the cost structure of large model companies, training costs occur before the model is released and are a one-time but enormous computing investment; inference costs, by contrast, occur continuously with user call volume, and the more users there are and the more frequent the calls, the greater the expenditure.76% similarUnverified当前多模态模型训练面临视觉模型和语言模型参数量相差数百倍的挑战75% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/489065API
curl https://kongchang.com/api/v1/knowledge/claims/489065MCP
get_claim(id=489065)