Very large: 300 billion to 1 trillion
PaLM, Llama 3.1 405B, DeepSeek-V3 and the big mixture-of-experts models. 37 models.
WorldMiniMax-M3
Native multimodal model with 1M context and sparse attentionMultimodal · Open weights
🇺🇸 United StatesTrinity Large
398B-parameter sparse MoE model family by Arcee AIGeneral · Open weights · 2026
🇨🇳 ChinaGLM-5
Open-weight model for engineering and agentic tasksGeneral · Open weights · 2026
🇨🇳 ChinaHunyuan-TurboS
Hybrid Mamba-Transformer MoE model by TencentGeneral · API only · 2025