maternion/ lfm2.5

947 3 days ago

LFM2.5-8B-A1B is a fast, memory-efficient hybrid MoE language model, designed for high-quality on-device AI and edge deployment. Unsloth Dynamic quants available!

tools thinking 8b

10 models

lfm2.5:latest
eb602b24a1f1 • 5.2GB • 125K context window •
Text input • 3 days ago
lfm2.5:8b
4214ff9627cf • 5.2GB • 125K context window •
Text input • 3 days ago
lfm2.5:8b-Q2_K_XL
3c6dc807d061 • 2.9GB • 125K context window •
Text input • 3 days ago
lfm2.5:8b-Q3_K_XL
d9763bf936a6 • 4.0GB • 125K context window •
Text input • 3 days ago
lfm2.5:8b-Q4_K_XL
186bbea02132 • 5.4GB • 125K context window •
Text input • 3 days ago
lfm2.5:8b-Q6_K_XL
6092647a8c60 • 7.7GB • 125K context window •
Text input • 3 days ago
lfm2.5:8b-Q8_K_XL
0ab4a29671f7 • 9.3GB • 125K context window •
Text input • 3 days ago
lfm2.5:8b-q4_K_M
ea6e8d18e20c • 5.2GB • 125K context window •
Text input • 3 days ago
lfm2.5:8b-q8_0
e1fa9350089e • 9.0GB • 125K context window •
Text input • 3 days ago
lfm2.5:8b-IQ2_M
ea2f8127f2e3 • 2.8GB • 125K context window •
Text input • 3 days ago