Expired70% confidenceFactExact time
DeepSeek's inference cost per unit of performance is approximately 1/5 to 1/10 that of dense models at the same tier.
2
Sources
70%
Confidence
Medium-term (~90 days)
Relevance
7/8/2026
First Seen
Valid until: 10/6/2026(expired)
Sources
Claude Code桌面版完整指南：无账号接入DeepSeek国产模型
bilibili程序员比巴布7/6/2026
Related Claims
UnverifiedDeepSeek Pro's 1.6T total parameters with 49B activated parameters result in inference costs comparable to a roughly 49B dense model, while expanding knowledge capacity by approximately 32 times.79% similarUnverifiedDeepSeek replicated top-tier model performance at extremely low compute costs.77% similarUnverifiedDeepSeek reported selected dense GEMM results approaching the hardware upper limits they claimed, measured on Ascend 950D with CANN 9.20 using L2 profiling, but these are DeepSeek's own non-independently reproduced numbers.74% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/260300API
curl https://kongchang.com/api/v1/knowledge/claims/260300MCP
get_claim(id=260300)