Expired75% confidenceFactTime unknown
The model context length currently invoked by Cursor is only 200,000 tokens.
1
Sources
75%
Confidence
Medium-term (~90 days)
Relevance
6/1/2026
First Seen
Valid until: 8/30/2026(expired)
Sources
Related Entities
Related Claims
UnverifiedTiny-MoE模型总参数量约200M，但每个token激活的参数仅约90M69% similarUnverified对于 198B 总参数的稀疏 MoE 模型，实际激活参数量可能仅为 20-40B 量级66% similarUnverified在SWE-bench Full版本上，多数模型的通过率仍低于20%66% similarUnverifiedThe 34B 80-layer model increases inference time by approximately 18% compared to the 27B65% similarUnverifiedA model with 100 billion parameters requires approximately 80 GB of VRAM to process a single inference request, while a model with tens of billions of parameters requires only about 20 GB.63% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/25295API
curl https://kongchang.com/api/v1/knowledge/claims/25295MCP
get_claim(id=25295)