Shanghai's Endless Frontier Lab Releases BigBang-v1, a 36B Self-Evolving LLM
AdinaYakup · x · 2026-08-07
Endless frontier lab based in Shanghai has released BigBang-v1, a 36B parameter large language model. It features a self-evolving training approach, learning from AI-generated frontier tasks. The lab claims that with only 10K high-quality examples, the model matches the performance of much larger models in science, coding, tool use, and long-context reasoning.
More from Models
- WeirdML v2 Benchmark Released: Tasks Expanded to 19, Clear Cost-Performance Scaling — yacineMTB · 2026-08-07
- MameLoshnLM: First Open-Source 8B LLM and Benchmark for Yiddish — Yiddish-NLP · 2026-08-07
- New Team Noiz Unveils Real-Time Interactive Multimodal Model at 1% of Mainstream Inference Cost — 新智元 · 2026-08-07
- Wired: One of China's Most Powerful AI Models, Kimi K3, Escapes Sandbox — wiredmagazine · 2026-08-07
- DeepSeek Flash Hits 80% of GPT 5.6 Luna Quality at 1/6 the Cost — pbaylies · 2026-08-07
- GPT 5.6 Luna Price Cut by 80%, AI Analytics Cost Drops Under Half a Cent — JeremyCMorgan · 2026-08-07