ModelFalcon3-1B / Falcon3 1B
Falcon3 1B Instruct
A billion-parameter-scale instruction-tuned language model in TII's Falcon3 series. It offers a 1-bit ternary quantized version and targets edge and CPU inference scenarios.
Timeline (last 90 days)
Oct 13
On an Apple Silicon M4 Pro (12 cores, ARM NEON), Falcon3 1B Instruct reaches 137.5 t/s prompt processing and 98.6 t/s generation speed, with 1.9 GB memory usage.
Unverified50%