Unverified90% confidenceFactTime unknown
Meta发布的Llama 3.1已支持405B参数规模，并以宽松的商业授权开放
1
Sources
90%
Confidence
Long-term
Relevance
5/31/2026
First Seen
Sources
Related Entities
Related Claims
UnverifiedMeta's Llama 3.1 comes in multiple parameter versions ranging from 8B to 405B, covering full-scenario needs from edge deployment to data centers.79% similarUnverifiedMeta's Llama 3.2 series introduced lightweight versions of 1B and 3B parameters.66% similarUnverifiedMeta released a local large language model called Muse Glimmer 30B, licensed under Apache 2.0, with 29.6 billion parameters, and it is a dense model.66% similarUnverifiedMeta's LLaMA 3 series has a maximum scale of 405B, and DeepSeek-V3 is approximately 671B (mixture-of-experts architecture).64% similarUnverifiedMiMo VL Pro总参数量达1T，激活参数42B，支持100万上下文窗口，采用混合专家（MoE）架构63% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/11869API
curl https://kongchang.com/api/v1/knowledge/claims/11869MCP
get_claim(id=11869)