OpenRouter· @OpenRouter · X·· 1 小时前AI 评分55
AI 导读
LiquidAI 发布 LFM2.5-24B-A2B，这是 LFM2.5 家族最大成员，已上线 OpenRouter。该模型为 24B 总参数、每 token 2B 激活的 MoE 推理模型，上下文扩展到 131K tokens，词表 128K，使用 33T 预训练 tokens，面向笔记本和 PC 本地运行多步 Agent 与工具调用任务；官方称在 Agent 和工具使用任务上可与总参数和激活参数多约 50% 的 Qwen3.6-35B-A3B 竞争，开放 base 和 post-trained 权重于 Hugging Face。
正文
LFM2.5-24B-A2B from @liquidai is now on OpenRouter.
A 24B MoE with 2B active parameters that reasons before it answers, built for agents and tool use.
Today we release 𝙻𝙵𝙼𝟸.𝟻-𝟸𝟺𝙱-𝙰𝟸𝙱, the largest 𝙻𝙵𝙼𝟸.𝟻 family member. It is an efficient local model for laptops and PCs to work through multi-step tasks inside popular agent harnesses. > 24B total parameters, 2B active per token > Context expanded to 131K tokens > A 128K-token vocabulary > 33T pretraining tokens > It is a reasoning model > Very easy to fine-tune and customize > Performs competitively with Qwen3.6-35B-A3B, a model with ~50% more total and active parameters, on agentic and tool-use tasks. > Open base and post-trained weights on @huggingface 🤗 1/在 X 查看被引用的帖子
来源：OpenRouter · x.com