Expired50% confidenceFactExact time
Chinese open-source teams such as DeepSeek, Qwen, and GLM have trained globally competitive models using restricted hardware against the backdrop of continuously tightening U.S. export controls on AI chips to China.
1
Sources
50%
Confidence
Medium-term (~90 days)
Relevance
7/13/2026
First Seen
Valid until: 10/11/2026(expired)
Sources
开源AI还剩6个月？深度解析开源大模型的生存危机与出路
Related Claims
UnverifiedDeepSeek's previous models were trained using Nvidia H800 chips before export restrictions were implemented.66% similarUnverifiedDeepSeek早期模型使用英伟达H800芯片训练完成，H800是英伟达专为中国市场削减互联带宽后推出的H100替代版本64% similarUnverifiedDeepSeek and Peking University jointly open-sourced the DSpark inference framework, which focuses on speculative decoding technology, improving inference speed by 85% and generation speed by 6–8% without replacing model weights.63% similarUnverifiedDeepSeek的R1模型是在美国禁售前用英伟达H800芯片训练的63% similarUnverifiedDeepSpeed and Megatron-LM are mainstream large model training frameworks.63% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/502883API
curl https://kongchang.com/api/v1/knowledge/claims/502883MCP
get_claim(id=502883)