AIWeaviate 1.39 adds 4-bit Rotational Quantization
Weaviate says the new option reduces memory use while maintaining strong recall, alongside faster imports and search performance improvements across its Rotational Quantization family.
TLDR
Weaviate has announced 4-bit Rotational Quantization (RQ) in version 1.39, saying it cuts memory use while maintaining strong recall. The company also reports faster imports and search improvements across the RQ family. In Weaviate Cloud, 8-bit RQ remains the default and now imports 16% faster, according to Weaviate.
Combined views
24
1 Source, first seen 13h ago
Weaviate 1.39 adds 4-bit Rotational Quantization
Weaviate says the new option reduces memory use while maintaining strong recall, alongside faster imports and search performance improvements across its Rotational Quantization family.
TLDR
Weaviate has announced 4-bit Rotational Quantization (RQ) in version 1.39, saying it cuts memory use while maintaining strong recall. The company also reports faster imports and search improvements across the RQ family. In Weaviate Cloud, 8-bit RQ remains the default and now imports 16% faster, according to Weaviate.