MedAIBase has released AntAngelMed, a 103B-parameter open-source medical language model that uses a 1/32 activation-ratio Mixture-of-Experts (MoE) architecture to activate only 6.1B parameters at inference time, matching the performance of roughly 40B dense models while exceeding 200 tokens per second on H20 hardware. Built on Ling-flash-2.0 and trained through a three-stage pipeline of continual pre-training, supervised fine-tuning, and GRPO-based reinforcement learning, the model ranks first among open-source models on OpenAI's HealthBench and tops both MedAIBench and MedBench leaderboards.
The post Meet AntAngelMed: A 103B-Parameter Open-Source Medical Language Model Built on a 1/32 Activation-Ratio MoE Architecture appeared first on MarkTechPost.
SOCIAL SHARE CARD GENERATOR