[DeepSeek's New Model MODEL1 Code Revealed, Suspected to Be a Brand-New Architecture] According to reports from Quantum Bit, the name 'MODEL1' has appeared for the first time in DeepSeek's updated FlashMLA code on GitHub, mentioned 28 times across 114 files. It is listed alongside the current version V32 (DeepSeek-V3.2), suggesting that MODEL1 represents the next-generation architecture model. Code differences indicate that this model has been optimized in areas such as KV cache layout, sparsity handling, and FP8 decoding, and it is expected to be officially released around the Lunar New Year. Combined with the recently disclosed mHC residual connection mechanism and Engram memory module, MODEL1 is anticipated to integrate multiple self-developed innovations.