MobileLLM-R1-140M-base: benchmark results
Meta MobileLLM-R1 140M Base is the pretrained checkpoint for downstream adaptation, separate from its final reasoning SFT release. The publisher specifies 15 layers, 9 attention heads, 3 key/value heads and shared input/output embeddings. Provider: Meta. Released 2025-09-12. Access: Open.
Aggregate Rating 1192 ± 36, rank #9659 of 10916 rated models, from 143 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| EuroEval Lithuanian Common Sense Reasoning | 0.42 | Common Sense Reasoning Average Score (%) | 25.1 |
| EuroEval Czech NLU - CS GEC | 2.87 | Matthews correlation coefficient (%) | 24.6 |
| EuroEval Latvian NLU - ScaLA LV | 0.62 | Matthews correlation coefficient (%) | 23.7 |
| EuroEval Faroese NLU - ScaLA FO | 0.15 | Matthews correlation coefficient (%) | 20.5 |
| EuroEval Icelandic Common Sense Reasoning | 0.75 | Common Sense Reasoning Average Score (%) | 20.4 |
| EuroEval Norwegian NLU - ScaLA NN | 1.08 | Matthews correlation coefficient (%) | 18.9 |
| EuroEval Norwegian NLU - ScaLA NB | 1.34 | Matthews correlation coefficient (%) | 18.3 |
| EuroEval Danish Knowledge - Danish Citizen Tests | 8.33 | MCC (x100) | 18.1 |
| EuroEval Danish NLU - ScaLA DA | 0.51 | Matthews correlation coefficient (%) | 17.9 |
| EuroEval English NLU - SQuAD | 45.31 | Character F1 (%) | 17.8 |
| EuroEval Dutch NLU - ScaLA NL | 2.25 | Matthews correlation coefficient (%) | 17.5 |
| EuroEval Icelandic NLU - ScaLA IS | 0 | Matthews correlation coefficient (%) | 17.4 |
Interactive version: theaggregate.ai/model?slug=mobilellm-r1-140m-base · How It Works · Data refreshed daily, snapshot 2026-11-08.