ποΈ Busy until day 13
AxionLab
AxionLab-official
AI & ML interests
Owner of SupraLabs and iGPU Lover
Recent Activity
liked a model 2 minutes ago
DinoResearch/Dino-50-v1 new activity 3 minutes ago
AxionLab-official/RizzAura-100k:Ngl, making a joke model soon π reacted to Banaxi-Tech's post with π§ about 1 hour ago
Introducing BananaMind 2 Nano
BananaMind 2 Nano is the smallest member of the BananaMind 2.0 family β a 10M-parameter language model that shows how much you can squeeze out of a tiny footprint. It uses the family's digit-isolated tokenizer, so it keeps solid arithmetic despite its size, and it's small enough to run just about anywhere.
Trained on 30B tokens in about a day on a single RTX 5070 Ti (16GB), 4096-token context.
Benchmarks:
Average 35.77
ARC Easy 36.20
PIQA 55.98
ARC Challenge 23.38
HellaSwag 27.50
That 35.77 average edges out Pythia-31M (~34.79) at roughly a third the parameters.
Released under Apache 2.0 on Hugging Face: https://huggingface.co/BananaMind/BananaMind-2-Nano β weights, tokenizer, and config included.