lthn/LEM-Gemma3-1B

Text Generation • 1B • Updated 9 days ago • 894 • 1

Note Foundation teacher — trained with Ethics-Composure-Ethics sandwich. Beats unaligned 27B on grammar.

lthn/LEM-Gemma3-4B

Text Generation • 0.6B • Updated 9 days ago • 763 • 1

Note 25th globally for IF on LiveBench. Trained via CL-BPL cascade from 1B teacher.

Note 1,189 grammar scores + 35 A/B test files across 29 models.

Note Paper, analysis, model cards. Section 13: CL-BPL cascade + LiveBench validation.

lthn/LEM-Gemma3-4B-GGUF

Text Generation • 4B • Updated 9 days ago • 1.15k

Note 17 GGUF quants (1-bit to 16-bit) for llama.cpp, ROCm, CPU.

lthn/LEM-Gemma3-1B-GGUF

Text Generation • 1B • Updated 9 days ago • 565

Note 12 GGUF quants (3-bit to 16-bit) for llama.cpp, ROCm, CPU.