Unverified50% confidenceFactExact time
LEANN compressed a 60-million-chunk (60M-chunk) RAG index from 201 GB to 6 GB.
1
Sources
50%
Confidence
Long-term
Relevance
10/28/2026
First Seen
Sources
Related Entities
Related Claims
UnverifiedConverting the 8-bit Ngram table to 4-bit can save approximately 100 GB, reducing the model size from approximately 475 GB to approximately 385 GB.72% similarVerified4-bit quantization compresses each parameter to 0.5 bytes, theoretically compressing 140GB down to approximately 35–40GB.71% similarUnverifiedQuantized model files are generally between 6 GB and 20 GB in size.71% similarUnverifiedThe DeepSeek 8B model compresses the model file to approximately 4.7 GB through 4-bit quantization.69% similarUnverifiedAt the 100 million token scale, the archive occupies 28.8 GB plus a 2.2 GB index on disk, while in-process memory via memory-map is only approximately 28 MB.68% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/1105694API
curl https://kongchang.com/api/v1/knowledge/claims/1105694MCP
get_claim(id=1105694)