tanaos/artifex

Small Language Model Inference, Fine-Tuning and Observability.

101
Stars
15
Forks
32.9
Radar Score
Stable
Momentum

Scoring Breakdown

Impact32.6
Velocity31.8
Health34.7

Growth Metrics

7-Day Stars
+0
30-Day Stars
+0
Acceleration
x
Language
Python

Why it ranks

Modest but positive momentum.

LLM Frameworks ยท inference engineTop 76% globally
View on GitHub