largha
4 TagsUpdated Apache-2.0by Phocinae
Phocinae's Largha: Laya's decision network on mmBERT-small (144M), fine-tuned on the typed-decisions workflows. Small enough for a CPU, on the Apple GPU through MLX.
multilingualfine-tuned144m
Tags
| Name |
|---|
| larghalatestSame as largha:150m.328 MB · 512 ctx · 100+ languages |
| largha:150mLargha v1.1 (mmBERT-small, 144M): 0.398 on Bespoke Labs' public benchmark; 18 ms for five questions on an RTX 5090.328 MB · 512 ctx · 100+ languages |
| largha:150m-fp16fp16 variant of largha:150m328 MB · 512 ctx · 100+ languages |
| largha:150m-fp32fp32 variant of largha:150m328 MB · 512 ctx · 100+ languages |
Tags without a suffix load fp16 on a CUDA GPU and fp32 on CPU; add -fp16 or -fp32 to pin one precision.