Unverified50% confidenceEventExact time
部分用户询问Opus 4.8是什么模型时，它有时自称Qwen或提到DeepSeek，引发关于模型蒸馏或训练残留的猜测
1
Sources
50%
Confidence
Medium-term (~90 days)
Relevance
7/2/2026
First Seen
Valid until: 9/30/2026
Sources
Claude Opus 4.8深度解析：越诚实越会应试的AI悖论
bilibili皮皮蟹勇闯天涯6/1/2026
Related Claims
UnverifiedDuring API testing, Claude Opus 4.8 incorrectly identified itself as Qwen and DeepSeek when asked what model it was74% similarUnverified有推测认为DeepSeek V4可能蒸馏了闭源专有模型，因其输出有时与闭源旗舰模型惊人相似69% similarUnverifiedDeepSeek-R1和Qwen-QwQ等开源模型在积极探索类似o1的慢思考推理方向67% similarUnverifiedDeepSeek-V3、Qwen等国内模型通过模型蒸馏技术，在参数量仅为OpenAI顶级模型数分之一的情况下实现了相当比例的性能复现66% similarUnverifiedQwen、Mixtral、DeepSeek等开源大模型系列均采用MoE设计65% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/48706API
curl https://kongchang.com/api/v1/knowledge/claims/48706MCP
get_claim(id=48706)