Subscribe and Support
This is LFM2.5-1.2B-Thinking quantized with AutoRound to GGUF, mixing different precisions.
Several versions (higher bpw is better and bigger):
- 3 average bits per weight (bpw)
- 4 bpw
- 5 bpw
Tested with LM Studio.
- Developed by: The Kaitchup
- License: lfm1.0