kanana-2-1.3B-base: benchmark results

Provider: Kakao. Access: Open.

Aggregate Rating 1352 ± 34, rank #7649 of 10552 rated models, from 26 benchmark results.

Strongest benchmark results

BenchmarkScoreMetricPercentile
Lara Mini Base Report (base-model protocol) - MBPP (3-shot)47.6pass@1 (execution) (%)64.8
Lara Mini Base Report (base-model protocol) - MBPP+ (3-shot)51.6pass@1; must pass MBPP's own and the EvalPlus tests (%)62
Lara Mini Base Report (base-model protocol) - PAWS (English) (0-shot)65.5Accuracy (%), English benchmark58.3
Lara Mini Base Report (base-model protocol) - CRUXEval-O (2-shot)32.5pass@1, direct prediction, official answer parser (%)50.9
Lara Mini Base Report (base-model protocol) - MATH (4-shot)31.4Math-Verify equivalence, Minerva prompts, 5,000 problems ove50
Lara Mini Base Report (base-model protocol) - MMLU-Pro (5 (CoT)-shot)29.9Exact match (%), English benchmark50
Lara Mini Base Report (base-model protocol) - MMLU (5-shot)56.4Accuracy (%), English benchmark44.4
Lara Mini Base Report (base-model protocol) - MGSM (8 (CoT)-shot)25.7Exact match on each language's native CoT answer phrase, ave42.6
Lara Mini Base Report (base-model protocol) - PIQA (0-shot)76Accuracy normalized by answer length (%), English benchmark41.7
Lara Mini Base Report (base-model protocol) - Belebele (English) (0-shot)60.4Accuracy (%), English benchmark38.9
Lara Mini Base Report (base-model protocol) - Global MMLU, All Languages (5-shot)38.9Accuracy (%) averaged over all languages/variants of the ben38.9
Lara Mini Base Report (base-model protocol) - Belebele, All Languages (0-shot)30.5Accuracy (%) averaged over all languages/variants of the ben38

Interactive version: theaggregate.ai/model?slug=kanana-2-1-3b-base · How It Works · Data refreshed daily, snapshot 2026-11-16.