Unverified60% confidenceFactExact time
DistilBERT compresses BERT's size by 40% while retaining 97% of its performance.
2
Sources
60%
Confidence
Long-term
Relevance
7/20/2026
First Seen
Sources
Related Claims
UnverifiedDistilBERT compressed BERT-base from 110M parameters to 66M, improving inference speed by 60% while retaining 97% of performance.79% similarUnverifiedDistilBERT retains approximately 97% of the original model's performance on the GLUE benchmark, with inference speed improved by approximately 60% and model size reduced by 40%.78% similarVerifiedDistilBERT reduces BERT's parameter count by 40% through distillation, increases inference speed by 60%, and incurs only approximately 3% performance loss70% similarUnverifiedDistilRoBERTa是RoBERTa的蒸馏版本，参数量减少40%但保留97%的性能70% similarVerifiedModel distillation can compress parameter count to 1/10 of the original while retaining 70–80% of the original performance.69% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/575453API
curl https://kongchang.com/api/v1/knowledge/claims/575453MCP
get_claim(id=575453)