PANews August 21 news, DeepSeek mentioned in its official announcement article "Officially Released DeepSeek-V3.1" that DeepSeek-V3.1 uses UE8M0 FP8 Scale parameter precision. In addition, V3.1 has made significant adjustments to the tokenizer and chat template, showing clear differences from DeepSeek-V3. DeepSeek's official WeChat account stated in a pinned message that UE8M0 FP8 is designed for the upcoming next generation of domestic chips.