SAVRN
Search Contact SAVRN

Organization

GREENBITAI

GreenBitAI

Low-bit neural networks, Edge AI, sustainable ML

Models in Library1
Datasets in Library0
Models on Hugging Face306
Followers37

Models

Model · Image and text to text

Qwen3.8-Flash-Next-4bit-paged

GREENBITAI

Expert-paged build of Vontra/Qwen3.8-Flash-Next-MLX-4bit. The weights that are read a fraction at a time live in their own containers, so a machine loads what it needs rather than all Total 105.46 GiB. Of that, 103.94 GiB is the source build, whose bytes moved into containers rather than being copied, and 1.52 GiB is the draft head, which no published build of this model carries. Where the weights fit they are filled from experts.bin and the model runs the stock path at stock speed; where they do not, they stream from disk. Reading the machine decides that, not a flag. To override that: GBXPAGING=off holds the experts resident, GBXPLE=off holds the n-gram table resident. Checked at build…

Open weights other 5.4B parameters 262,144 tokens mlx