Cointime reports: CCID Consulting research published in its English-language journal Foresight on November 6 found that generative AI models above 100 billion parameters achieve only a 5% return on investment, while domain-specific models in the 7-30 billion parameter range reach 35%. Performance plateaus at 40-60 billion parameters in structured commercial scenarios, and inference accounts for 70%-90% of total lifecycle costs, the report said.
Large Models Over 100B Parameters Return Only 5%, CCID Consulting Says
Share on WeChat
Open WeChat → Scan → then tap "…" to send to a chat or Moments.
Tap "…" in the top-right corner to send to a chat or share to Moments.
Related News
8m ago
SanDisk shares rebound over 11% in premarket after 22% plunge
37m ago
MiniMax CEO: Open Source Until AGI, Profit via Lower Inference Costs
1h ago
US Group Finds Anthropic AI Model Capability Gains Accelerating
1h ago
House Leaders Ask Nvidia CEO Jensen Huang to Brief on Megaspeed Executive Ties
1h ago
Singapore’s Symatics unveils MPU AI chip architecture
1h ago
Cipher Takes On $359M Cost Overrun at Barber Lake Data Center