BigBang-V1: 35B Model Beats 1.6T Parameter Giants

A Chinese research team introduced BigBang-V1, a 35B parameter model trained natively on Recursive Self-Improvement (RSI) with 100% self-generated synthetic data, successfully outperforming 1.6 trillion parameter models in several scientific benchmarks.

2026-08-06 ~ 2026-08-07 · 3 related posts

• Chinese Team Debuts BigBang-V1: A 35B Native RSI Model Beating DeepSeek V4 — 新智元 · 2026-08-06
• 35B Beats 1.6T: BigBang Model Achieves SOTA via Self-Evolving Synthetic Data — ChengleiSi · 2026-08-07
• 35B Model Beats 1T: BigBang Achieves Self-Evolution with 100% AI-Synthesized Data — 量子位 · 2026-08-07