StentorLabs high quality datasets for training language models.
Kai Izumoto
StentorLabs
AI & ML interests
I'm Kai Izumoto, an independent AI researcher and developer building compact, efficient language models through StentorLabs. I'm passionate about democratizing AI research by demonstrating what's possible on free-tier hardware and accessible compute.
Recent Activity
new activity
1 day ago
stas122/stentor_python_30m:🔥🔥🔥 updated
a collection
1 day ago
Datasets upvoted a collection 1 day ago
Datasets Organizations
None yet
Independent, transparent research into small and large language models .
The Stentor1 series is StentorLabs’ first family of transparent 12M- and 30M-parameter Llama-style models trained from scratch on free-tier hardware.
-
StentorLabs/Stentor-30M
Text Generation • 30.4M • Updated • 296 • 2 -
StentorLabs/Stentor-12M
Text Generation • 12M • Updated • 178 • 1 -
StentorLabs/Stentor-30M-Instruct
Text Generation • 30.4M • Updated • 158 • 1 -
StentorLabs/Stentor-12M-Instruct
Text Generation • 12M • Updated • 27
StentorLabs high quality datasets for training language models.
Independent, transparent research into small and large language models .
The Stentor2 series is an improved version of the Stentor1 series, featuring refinements while keeping full transparency.
The Stentor1 series is StentorLabs’ first family of transparent 12M- and 30M-parameter Llama-style models trained from scratch on free-tier hardware.
-
StentorLabs/Stentor-30M
Text Generation • 30.4M • Updated • 296 • 2 -
StentorLabs/Stentor-12M
Text Generation • 12M • Updated • 178 • 1 -
StentorLabs/Stentor-30M-Instruct
Text Generation • 30.4M • Updated • 158 • 1 -
StentorLabs/Stentor-12M-Instruct
Text Generation • 12M • Updated • 27