Expired50% confidenceFactExact time
Inference throughput on equivalent hardware has improved several-fold to tens-of-fold over the past two years, while market pricing has declined at a slower pace.
1
Sources
50%
Confidence
Medium-term (~90 days)
Relevance
7/10/2026
First Seen
This claim has expired. It is kept for the record and no longer describes the current state.
Valid until: 10/8/2026(expired)
Sources
AI推理服务真的盈利吗？拆解推理与训练的成本经济学
hackernewshackernews7/3/2026
Related Claims
UnverifiedIndustry estimates suggest that the inference cost of mainstream LLMs drops approximately 75% every 18 months, but demand is growing faster, causing total costs to continue rising68% similarUnverifiedThe decline curve of AI inference costs over the next few years will be steeper than most people expect.64% similarUnverifiedAI inference prices, measured per unit of compute, have dropped approximately 10-fold over the past two years, a phenomenon the industry calls the AI inference price spiral.64% similarUnverifiedThe NVIDIA H200 improves inference efficiency by approximately 3x compared to the A100 on specific tasks, but the simultaneous expansion of model parameter scale has almost offset the cost dividend brought by hardware progress.63% similarUnverifiedClaude Code's fast mode is 2.5 times faster, costs approximately three times less than before, and token consumption when processing PDFs and charts is 61% lower than the previous generation.63% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/412210API
curl https://kongchang.com/api/v1/knowledge/claims/412210MCP
get_claim(id=412210)