Expired50% confidenceFactExact time
Microsoft is betting on the topological qubit route, attempting to reduce error-correction overhead by constructing Majorana fermions.
1
Sources
50%
Confidence
Medium-term (~90 days)
Relevance
7/7/2026
First Seen
Valid until: 10/5/2026(expired)
Sources
后量子密码学Python实战：ML-KEM密钥封装机制详解
hackernewshackernews7/4/2026
Related Claims
UnverifiedSiliconFlow's underlying system uses a self-developed inference acceleration engine supporting optimization technologies such as FlashAttention and continuous batching, which can reduce inference costs by more than 50% compared to native deployment.63% similarUnverified微软研究院于2023年率先提出BitNet架构，其论文证明在Scaling Law作用下1-bit模型的精度损失会随参数量增大而逐步收窄61% similarUnverified微软研究院于2023年率先提出BitNet架构，证明1-bit网络在保持合理精度的同时能大幅降低计算复杂度60% similarUnverifiedOllama 底层基于 llama.cpp 的量化推理引擎，通过 4-bit/8-bit 量化技术降低显存占用59% similarUnverifiedThe surface code, owing to its approximately 1% fault-tolerance threshold and regular two-dimensional lattice topology, has become the core error-correction scheme in the mainstream quantum hardware roadmaps of Google and IBM.58% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/196387API
curl https://kongchang.com/api/v1/knowledge/claims/196387MCP
get_claim(id=196387)