eoba

@eoba

Joined February 2026

Projects

nemotron-3-nano-4b
Public
Forked from nvidia/nemotron-3-nano-4b

General purpose reasoning and chat model trained by NVIDIA

MODEL

1

Updated on March 18
granite-4-h-tiny
Public
Forked from ibm/granite-4-h-tiny

A hybrid MoE model trained for tool use from IBM.

MODEL

2

Updated on February 21
functiongemma-270m
Public
Forked from google/functiongemma-270m

Lightweight Gemma 3-based model (270M params) trained specifically for function calling. Text-only with a 32k context window, designed to be fine-tuned into your own tool agent while remaining small enough for laptops or edge devices.

MODEL

1

Updated on February 21