MoE models around A2B
英文标题：MoE models around A2B
There's a bunch of small MoE with around 1B active params, like LFM2.5 8B A1B and Granite 4.0h 7B A1B; and then there are models with 3B+ like Qwen 3.x ~30B A3B and Gemma 4 26B A4B, but those are already on the heavier side if you don't have enough resources. What about the middle ground, MoE with a
打开原文相关 2 · 机会 0
中文
There's a bunch of small MoE with around 1B active params, like LFM2.5 8B A1B and Granite 4.0h 7B A1B; and then there are models with 3B+ like Qwen 3.x ~30B A3B and Gemma 4 26B A4B, but those are already on the heavier side if you don't have enough resources. What about the middle ground, MoE with a
中文全文翻译正在补齐，可先查看下方英文原文。
English / 原文
There's a bunch of small MoE with around 1B active params, like LFM2.5 8B A1B and Granite 4.0h 7B A1B; and then there are models with 3B+ like Qwen 3.x ~30B A3B and Gemma 4 26B A4B, but those are already on the heavier side if you don't have enough resources. What about the middle ground, MoE with a