AI BriefingOpen SLM Evaluations Dataset Released with Benchmarks for 106 Small Language Models
·2026.10.08 05:49
Key point
The dataset covers 106 SLMs with 150M parameters or fewer, evaluated on a Runpod B200 under Apache 2.0 license.
Details
The Open SLM Evaluations dataset has been released, providing benchmark scores for 106 Small Language Models (SLMs) with 150M parameters and below. The data includes model names, parameter counts, links, and independent evaluation results.
Evaluation Details
- Scope: 106 SLMs (≤150M parameters).
- Hardware: Evaluated on a Runpod B200 GPU.
- Duration: Approximately 30 GPU minutes per evaluation run.
- License: Apache 2.0.
The dataset is available on Hugging Face, offering a standardized reference for comparing small-scale language model performance.
This summary was generated automatically by AI. Check the original for the author's claims and context. Copyright belongs to the original author.
Our guide explains how the AI works. Report summary errors, attribution issues, or removal requests via Contact.