LFM2.5-24B-A2B 开放权重，支持在本地运行推理型智能体Chinese only
LFM2.5-24B-A2B 及其基础版现已在 Hugging Face 提供开放权重，可用于在设备上运行会推理和调用工具的智能体。相较前代，其上下文窗口从 32K 扩至 131K tokens；发布方的测试显示，它在 LiveClaw 上得分 48.09，但在部分知识和编程评测中落后于更大的模型。
1 developments2 reports2 sources
Event developments 1
LFM2.5-24B-A2B open weights available for local reasoning agents
Open weights for LFM2.5-24B-A2B and its base version are now available on Hugging Face for agents that reason and call tools on-device. Its context window grows from 32K to 131K tokens over its predecessor. In the publisher's tests, it scored 48.09 on LiveClaw, while larger models retained an edge on some knowledge and coding benchmarks.
View source reports · 2 sources · 2 reports
- Hugging Face 官方博客 · LFM2.5-24B-A2B open weights available for local reasoning agents
- MarkTechPost AI 研究报道 · Liquid AI releases LFM2.5-24B-A2B, an agent model that runs on laptops
All reports 2
Liquid AI releases LFM2.5-24B-A2B, an agent model that runs on laptops
The open-weight model targets multi-step tool use, with base and post-trained weights available on Hugging Face. It has a 131,072-token context window. In Liquid AI’s tests, it scored 27.69 on AutomationBench but trailed a comparison model on SWE-Bench Verified, 51.00 to 66.20.
1 more source reported this
- Hugging Face 官方博客
LFM2.5-24B-A2B open weights available for local reasoning agents
Open weights for LFM2.5-24B-A2B and its base version are now available on Hugging Face for agents that reason and call tools on-device. Its context window grows from 32K to 131K tokens over its predecessor. In the publisher's tests, it scored 48.09 on LiveClaw, while larger models retained an edge on some knowledge and coding benchmarks.
1 more source reported this
- MarkTechPost AI 研究报道