I Trained an Open Decision Model on a Free GPU in 46 Minutes. On My Tests, It Beat Laya.

Description and use

Typic, trained in 46 minutes on a free GPU, outperformed Laya in accuracy and speed on unseen tasks, eliminating the need for text generation and parsing.

Problem

Large language models are slow, costly, and produce text that must be parsed for decision tasks.

Outcome

Typic achieved 69.9% accuracy on unseen tasks, outperforming Laya’s 56.7% and matching its intent routing accuracy, while running in 29 ms per question versus Laya’s 37 ms.

Workflow

• Collect 25 public datasets and convert each example to (context, question, options, answer) format.
• Apply extensive augmentation: vary question wording, label wording, option order, include "none of these", flip yes/no questions.
• Train Ettin‑400M encoder with a small head, using one T4 GPU, batch size 8, gradient accumulation, and checkpointing for 46 minutes.
• Fit a temperature parameter for calibration.
• Evaluate on 2,000 unseen questions across Banking77, CommonsenseQA, COPA, OpenBookQA, and long‑context routing tests.
• Compare results with Laya’s official package.
• Publish the model on Hugging Face.

Evidence

I trained three sizes along the way. On questions from tasks the model never saw during training: The tiny 68M pilot got only 33% on a five‑option test (random is 20%). 150M reached 58.3% across four unseen tasks. 400M reached 69.9%.

Source

Typic was 13 points more accurate overall, tied Laya on intent routing, and answered in 29 ms vs. 37 ms per question.

Source

I placed 20 support requests after 0 to 7,000 tokens of filler… Typic kept 17 or 18 out of 20 correct up to 7,000 tokens. Both Laya checkpoints dropped to 5 or 6 out of 20.

Source