Luciole-1B-Base: benchmark results

Provider: OpenLLM-France. Access: Open.

Aggregate Rating 1244 ± 32, rank #9156 of 10552 rated models, from 26 benchmark results.

Strongest benchmark results

BenchmarkScoreMetricPercentile
Lara Mini Base Report (base-model protocol) - XNLI (English) (0-shot)53.8Accuracy (%), English benchmark61.1
Lara Mini Base Report (base-model protocol) - Global PIQA (non-parallel), All Languages (0-shot)54Accuracy (%) averaged over all languages/variants of the ben33.3
Lara Mini Base Report (base-model protocol) - CRUXEval-I (2-shot)25.9pass@1, direct prediction, official answer parser (%)22.2
Lara Mini Base Report (base-model protocol) - HellaSwag (10-shot)62.5Accuracy normalized by answer length (%), English benchmark20.4
Lara Mini Base Report (base-model protocol) - PIQA (0-shot)73.6Accuracy normalized by answer length (%), English benchmark20.4
Lara Mini Base Report (base-model protocol) - WinoGrande (5-shot)61.7Accuracy (%), English benchmark20.4
Lara Mini Base Report (base-model protocol) - Belebele, All Languages (0-shot)27.9Accuracy (%) averaged over all languages/variants of the ben19.4
Lara Mini Base Report (base-model protocol) - PAWS-X, All Languages (0-shot)51.9Accuracy (%) averaged over all languages/variants of the ben19.4
Lara Mini Base Report (base-model protocol) - XNLI, All Languages (0-shot)37.6Accuracy (%) averaged over all languages/variants of the ben19.4
Lara Mini Base Report (base-model protocol) - COPA (0-shot)78.4Accuracy (%), English benchmark18.5
Lara Mini Base Report (base-model protocol) - MBPP (3-shot)22.6pass@1 (execution) (%)18.5
Lara Mini Base Report (base-model protocol) - T-HellaSwag, All Languages (0-shot)34.2Accuracy normalized by answer length (%) averaged over all l18.5

Interactive version: theaggregate.ai/model?slug=luciole-1b-base · How It Works · Data refreshed daily, snapshot 2026-11-16.