Moonlight-A3B - a moonshotai Collection

moonshotai 's Collections

Kimi-Linear-A3B

Kimi-K2

Moonlight-A3B

updated about 8 hours ago

Moonshot's Compute-efficient MoE LLM, first Scaling Up of Muon Optimizer

moonshotai/Moonlight-16B-A3B-Instruct

Text Generation • 16B • Updated Mar 3 • 24k • 184
moonshotai/Moonlight-16B-A3B

Text Generation • 16B • Updated Feb 26 • 15.2k • 95
Muon is Scalable for LLM Training

Paper • 2502.16982 • Published Feb 24 • 6