Expired60% confidenceFactExact time
Bonsai compresses a 27B model from 54GB at full precision down to 3.9GB
2
Sources
60%
Confidence
Medium-term (~90 days)
Relevance
7/20/2026
First Seen
Valid until: 10/18/2026(expired)
Sources
Related Claims
Unverified原版QW3.6 27B模型文件体积约为54GB，Prismo通过数学压缩技术压缩至不到4GB，压缩比超过13倍，且声称不影响模型性能80% similarUnverifiedBonsai 27B is the first 27B-class large model capable of running on a smartphone; through quantization technology, the 54 GB model is compressed to 3.8 GB—a size reduction of approximately 93%—while retaining approximately 90% of its intelligence level, and it can run locally on an iPhone 17 Pro.78% similarUnverifiedQ3_K_M compresses each weight to approximately 3 bits, reducing the weight size of a 27B model to approximately 12–13 GB.75% similarUnverifiedA 70B model can be compressed to approximately 35GB at INT4 precision.72% similarUnverifiedThe Bonsai ternary version is approximately 6GB at 1.71 bits and retains approximately 97% of the base model's quality; the 1-bit version is 3.9GB and retains approximately 90% of quality71% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/569975API
curl https://kongchang.com/api/v1/knowledge/claims/569975MCP
get_claim(id=569975)