Luth-LFM2-1.2B: benchmark results
Provider: Kurakura AI. Released 2025-08-22. Access: Open.
Aggregate Rating 1338 ± 30, rank #1386 of 1635 rated models, from 33 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| Kurakurai French LLM - GPQA Diamond | 27.4 | Accuracy (%) | 67.1 |
| Kurakurai French LLM - IFEval | 64.09 | Accuracy (%) | 67.1 |
| Kurakurai French LLM - MATH-500 | 46.28 | Accuracy (%) | 63.4 |
| Kurakurai French LLM - MGSM | 62.56 | Accuracy (%) | 63.4 |
| Kurakurai French LLM - MultiIF | 42.15 | Accuracy (%) | 63.4 |
| Kurakurai French LLM - Global MMLU | 51.2 | Accuracy (%) | 58.5 |
| CARTE - Pays de la Loire | 45.9 | Accuracy (%) on the 133 CARTE questions about Pays de la Loi | 57.7 |
| Kurakurai French LLM - HumanEval+ | 30.63 | pass@1 (%) | 56.1 |
| Kurakurai French LLM - MMLU-Pro | 26 | Accuracy (%) | 56.1 |
| Kurakurai French LLM - AIME 2025 | 0.67 | Accuracy (%) | 54.9 |
| CARTE-LV | 49.4 | Accuracy (%) on the 233 CARTE-LV questions on regional lingu | 53.8 |
| CARTE - Auvergne-Rhone-Alpes | 47.1 | Accuracy (%) on the 208 CARTE questions about Auvergne-Rhone | 50 |
Interactive version: theaggregate.ai/model?slug=luth-lfm2-1-2b · How It Works · Data refreshed daily, snapshot 2026-11-05.