Unverified85% confidenceFactTime unknown
DeepSeek V4 Pro的浮点运算量降至V3.2版本的27%（降低3.7倍）
1
Sources
85%
Confidence
Medium-term (~90 days)
Relevance
6/2/2026
First Seen
Valid until: 8/31/2026
Sources
DeepSeek V4技术深度拆解：百万Token与极致性价比
bilibili码书z
Related Entities
Related Claims
UnverifiedDeepSeek V4 Pro在CodeForces取得3206分，MMT-266评测达到95.2%通过率75% similarUnverifiedDeepSeek V4 Pro相比Qwen3.6 27B Q4领先约6%，如果不是在推理数学项上意外翻车，领先幅度可能超过10%73% similarUnverifiedLaguna S 2.1 宣称在编程能力上做到「14 倍小、4.5 倍强」，相对 DeepSeek V4 Pro Max 等大模型69% similarUnverified经过对比测试，DeepSeek V4 Pro的编程效果甚至超过了Claude Sonnet，而且价格便宜得多68% similarUnverifiedQwen3.5的参数激活率为4.3%，而DeepSeek V3的671B模型激活率为5.5%66% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/36597API
curl https://kongchang.com/api/v1/knowledge/claims/36597MCP
get_claim(id=36597)