MobileLLM-R1-360M-base: benchmark results
Meta MobileLLM-R1 360M Base is the pretrained checkpoint for downstream adaptation, separate from its final reasoning SFT release. The publisher specifies 15 layers, 16 attention heads, 4 key/value heads and shared input/output embeddings. Provider: Meta. Released 2025-09-12. Access: Open.
Aggregate Rating 1204 ± 35, rank #9592 of 10916 rated models, from 143 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| EuroEval Icelandic Common Sense Reasoning | 3.54 | Common Sense Reasoning Average Score (%) | 38.1 |
| EuroEval Lithuanian Common Sense Reasoning | 1.76 | Common Sense Reasoning Average Score (%) | 34.6 |
| EuroEval English NLU - SQuAD | 65.63 | Character F1 (%) | 27.6 |
| EuroEval Faroese NLU - FoSent | 4.33 | Matthews correlation coefficient (%) | 22.6 |
| EuroEval Norwegian NLU - ScaLA NN | 1.81 | Matthews correlation coefficient (%) | 22.5 |
| EuroEval Spanish NLU - MLQA ES | 33.88 | Character F1 (%) | 22.2 |
| EuroEval Icelandic Knowledge | 1.44 | Knowledge Average Score (%) | 22.1 |
| EuroEval Estonian Common Sense Reasoning | 0.58 | Common Sense Reasoning Average Score (%) | 20.6 |
| EuroEval French NLU - FQuAD | 28.53 | Character F1 (%) | 18.8 |
| EuroEval Italian NLU - ScaLA IT | 1.4 | Matthews correlation coefficient (%) | 18.2 |
| EuroEval Danish NLU - ScaLA DA | 0.47 | Matthews correlation coefficient (%) | 17.4 |
| EuroEval Icelandic NLU - ScaLA IS | 0 | Matthews correlation coefficient (%) | 17.4 |
Interactive version: theaggregate.ai/model?slug=mobilellm-r1-360m-base · How It Works · Data refreshed daily, snapshot 2026-11-08.