Sentence Generation on CommonGen benchmark leaderboard · SOTA2 Research

Sentence Generation on CommonGen

53.6Win-Tie ScoreGemma-3-27B-IT

Evaluation Results

Method	Links
Gemma-3-27B-IT2026.03		53.6	97.9	52.5	71.7	79.2	21.4	30.7
Qwen-2.5-14B-Inst.2026.03		52	96	49.9	73.2	80.6	22.1	32.6
Gemma-3-12B-IT2026.03		51.1	97.4	49.8	73.7	81.2	21.8	31.7
Qwen-2.5-7B-Inst.2026.03		48.4	93	45	73.7	81	22.5	33.7
Llama-3.1-8B-Inst2026.03		47.3	94.5	44.7	75.5	82.9	22.7	34.2
Qwen-2.5-14B-Inst.2026.03		45.2	95.9	43.4	42.4	47.5	16.5	18.6
Gemma-3-12B-IT2026.03		43.2	94	40.6	31.7	35.7	14.7	13.6
Qwen-2.5-7B-Inst.2026.03		42.9	92.3	39.6	39.2	44.1	15.8	16.3
Gemma-3-4B-IT2026.03		41.2	95.5	39.4	75.4	82.6	22.4	33.4
Gemma-3-27B-IT2026.03		40.3	95.6	38.5	70.7	79.2	20.3	28.1
Qwen-2.5-14B-Inst.2026.03		38.5	92.9	35.8	70.8	79.6	20.9	29.7
Gemma-3-12B-IT2026.03		35.9	94.1	33.8	72.2	80.4	20.5	28.7
Qwen-2.5-7B-Inst.2026.03		32.5	88.6	28.8	73.2	81.8	21.4	31.2
Gemma-3-27B-IT2026.03		31.9	86.7	27.7	27	30.2	15	14.7
Llama-3.1-8B-Inst2026.03		31.7	85.8	27.2	76.4	84.7	21.2	30.7
Gemma-3-4B-IT2026.03		28.7	88.4	25.3	37.7	42.3	16.1	17.4
Gemma-3-4B-IT2026.03		28.7	91	26.1	74.9	83.2	21.2	30.6
Llama-3.1-8B-Inst2026.03		19	84.7	16.1	73.4	80.3	22.8	35.1