Expired70% confidenceOpinionExact time
LFM2.5-8B-A1B在工具调用（tool calling）场景中官方声称可达到4倍体量模型的效果
1
Sources
70%
Confidence
Medium-term (~90 days)
Relevance
5/28/2026
First Seen
Valid until: 8/26/2026(expired)
Sources
Related Entities
Related Claims
UnverifiedLFM2.5在多项基准测试中展现出与体量高达其4倍模型相当的竞争力71% similarUnverifiedLFM2.5 improved by 74 points in a telecom test simulating real agent scenarios, achieved an instruction-following IFEval score of 91.8, and a mathematical reasoning Math500 score of 88.8.64% similarUnverifiedgemma-4-26b-a4b 模型名称中的 'a4b' 表示该模型采用了混合专家（MoE）架构，激活参数量约为4B64% similarUnverifiedLFM2.5-8B-A1B总参数量为8B，推理时仅激活1.5B参数63% similarUnverifiedV4-Pro uses a Mixture of Experts (MoE) architecture with a total parameter count of 1.6 trillion (1.6T) and 49 billion (49B) active parameters.63% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/626API
curl https://kongchang.com/api/v1/knowledge/claims/626MCP
get_claim(id=626)