Pollock Mini LM · 125M
A compact language-model release. Model card, weights and usage information are available in the public repository.
FABRYKA AI / PUBLICATIONS & RESEARCH OUTPUTS
Artifacts behind
the research.
Public models, datasets, evaluation infrastructure and technical documentation. Browse the original repositories for versions, licenses and usage information.
01 / MODELS & DATA
Inspect. Reuse.
Build on the work.
These links point to public research artifacts from SlayerLab. Repository documentation defines each artifact’s scope and limitations.
Model repository ↗
Polish DynaWord
A public Polish-language dataset supporting our work on language-model training and data composition.
GoLLeM v4 · 250M PL
A Polish language-model repository with published weights and model documentation.
02 / BENCHMARKS & TECHNICAL DOCUMENTATION
Make the comparison
inspectable.
Evaluation resources and practical documentation support the research workflow. They are listed by artifact type, separately from papers and technical reports.
CodeSOTA
Model comparisons and benchmark evidence. Individual entries identify their sources and evaluation context.
Token probabilities & distillation
API documentation and examples for collecting token probabilities and exploring distillation.
Slayer experiments & methodology
Public experiment write-ups and the laboratory’s evaluation methodology.
03 / PAPERS & TECHNICAL REPORTS
A place for
the complete study.
No papers or technical reports are listed in this catalogue yet. Forthcoming reports will link their authors, methodology, results, limitations and supporting artifacts.
Planned reports cover compact-model scaling and distillation, commodity-GPU inference, and Polish evaluation methods. These are research directions, not completed publications.
Explore the research agenda →