Unverified50% confidenceFactExact time
BitNet b1.58 restricts weights to the set {-1, 0, +1}, which is highly consistent with the approach taken by Bonsai 2.
1
Sources
50%
Confidence
Long-term
Relevance
10/4/2026
First Seen
Sources
Related Entities
Related Claims
VerifiedThe BitNet architecture proposed by Microsoft Research in 2023 is the theoretical pioneer of 1-bit large language models, with BitNet b1.58 restricting weights to ternary values (-1/0/+1).72% similarUnverified1-bit Bonsai retains most of its performance on math and coding tasks, but shows significant degradation in tool calling65% similarUnverifiedTernary quantization restricts weights to three values — -1, 0, and +1 — enabling addition and subtraction to replace multiplication in order to reduce hardware computational load, and is related to Microsoft's BitNet series of research.65% similarUnverifiedPrismML's flagship technology is extreme quantization, with its signature work being a 1-bit (ternary) model called bonsai, which compresses weights to three states: -1, 0, and +1.61% similarUnverifiedBitNet对激活值保留高精度(通常INT8)，仅对权重执行极端压缩，形成低精度权重加高精度激活的非对称量化60% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/973437API
curl https://kongchang.com/api/v1/knowledge/claims/973437MCP
get_claim(id=973437)