Research
ufakzeka-kulak: A Small Turkish Speech Recognition Model Trained from Scratch
An open Turkish speech recognition model of 31,564,935 parameters, trained from scratch, that runs offline on a CPU. The post covers its training data and method, the results of the release gate and its measured limits.
HakemBench: A Turkish Decision Benchmark
HakemBench holds 2,346 open Turkish items and 4,275 questions in seven tracks and scores models on decision quality, calibration and selective automation, with bootstrap intervals. This post explains what it measures and how its gold was made, and presents the leaderboard.
ufakzeka-karar: A Small Turkish Decision Model
An open Turkish decision model that answers typed questions about a text on a CPU, with a probability for every option and a not-sure signal. The post covers how it was built, where it stands on HakemBench and where it falls short.
ufakzeka-1: Building and Evaluating a 151M-Parameter Turkish Language Model from Scratch
The first model from ufak AI: pretrained from scratch on openly licensed data, fine-tuned for chat, benchmarked on one harness, limits measured, under $300 all in.