Unverified50% confidenceConceptExact time
General-purpose large models (such as GPT and Gemini) possess broad knowledge and reasoning capabilities, but their training and inference costs are extremely high.
1
Sources
50%
Confidence
Long-term
Relevance
11/3/2026
First Seen
Sources
Related Entities
Related Claims
UnverifiedIndustry estimates suggest that training a top-tier model once (such as GPT-4 or the Claude 3 series) often costs tens of millions to hundreds of millions of dollars in compute.73% similarUnverifiedWith the emergence of larger models such as GPT-4 and Gemini, training costs are considered to have risen by orders of magnitude.73% similarUnverifiedIn this context, frontier models refer to large-scale foundation models with highly general reasoning and code understanding capabilities, such as Gemini, the GPT series, and others, called via API with token-based billing.72% similarUnverifiedThe inference cost of large model services is mainly determined by GPU compute time, memory usage, and bandwidth.71% similarUnverifiedGPT-series models are optimized during training to maximize prediction accuracy (i.e., minimize perplexity), so their outputs tend toward high-probability word combinations.69% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/1131673API
curl https://kongchang.com/api/v1/knowledge/claims/1131673MCP
get_claim(id=1131673)