Unverified50% confidenceFactExact time
For models with parameter counts above 70B, a single machine or single node's VRAM is often unable to accommodate the full model, making Megatron-style model parallelism almost a mandatory choice.
1
Sources
50%
Confidence
Long-term
Relevance
9/11/2026
First Seen
Sources
Related Entities
Related Claims
UnverifiedWithin specific task domains, small- to medium-scale but highly specialized models often outperform general-purpose flagship models65% similarUnverifiedWhen model scale crosses certain thresholds, emergent capabilities not previously present arise (emergent abilities), such as multi-step logical reasoning and complex mathematical computation.65% similarUnverifiedLarge models may merely be absorbing the loss from compression using excess parameters and hidden state capacity, making the compression robustness observed on saturated tasks an illusion.65% similarUnverified通用大模型直接做专业仿真在单位制、参数规范等细节上极易出错，导致无效结果65% similarUnverified支持向量机的对偶问题通过拉格朗日对偶性转化后，计算复杂度只依赖于样本数而非特征维度，且数据仅以内积形式出现，为核技巧提供理论基础63% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/897659API
curl https://kongchang.com/api/v1/knowledge/claims/897659MCP
get_claim(id=897659)