GEHIRNS/seele-1.0-3.3B
Text Classification • 3B • Updated • 593 • 1
SEELE
Training-free pruned and quantized Jeffan-4B-2010 decision model. At 8K context: 98% of its score in 56% of the GPU memory, 1.4x faster.
Note Recommended. 1.70 GiB GGUF, ID 0.453 at 8K context (1.7x the Q8_0 parent), 6.5 decisions/s on an NVIDIA A16. Recipe and data: github.com/GEHIRNS/seele.