Our new X account is live! Follow @wizwand_team for updates

Grammaticality Judgment on RuBLiMP Russian 1.0

96.7Anaphora

Haiku (GPb+CT son)

Updated 4d ago

Evaluation Results

Method	Links
Haiku (GPb+CT son) 2025.06		96.7	74	100	94	100	97.3	100	94.6
gpt-4o (CT) 2025.06		96.7	87.3	96.7	95.9	86.7	96	100	94.2
gpt-4o (GPb+CT son) 2025.06		96.7	88	100	98.1	100	99	100	97.4
gpt-4o (GPb+CT ol) 2025.06		96.7	84	100	95.1	99.8	98	99.3	96.1
Sonnet (base) 2025.06		96	90	96	96.8	96.2	97.7	98.7	95.9
Haiku (GPb+CT ol) 2025.06		95.3	80.7	99.3	94.8	99.3	94	98	94.5
gpt-4o (GPb ol) 2025.06		95.3	90	100	96.5	98.9	97	100	96.8
Sonnet (GPb son) 2025.06		95.3	89.3	100	96.4	100	99.7	99.3	97.2
Llama (GPb+CT) 2025.06		95.3	90.3	100	92	95.1	94.9	87.8	93.6
gpt-4o (base) 2025.06		94.7	88.7	94.7	94.1	90.7	96.3	98.7	94
Sonnet (GPb ol) 2025.06		94.7	88	100	96.9	99.3	99	98.7	96.7
Sonnet (GPb+CT son) 2025.06		94.7	89.3	100	97.7	100	99	100	97.2
Sonnet (GPb+CT ol) 2025.06		94.7	86.7	100	98.1	99.3	98.7	97.3	96.4
Sonnet (CT) 2025.06		94	89.3	93.3	97.2	92.2	97.3	99.3	94.7
gpt-3.5 (GPb+CT ol) 2025.06		93.3	70	100	89.5	97.3	82.7	79.3	87.4
gpt-4o (GPb son) 2025.06		93.3	90.7	100	95.6	100	98.3	100	96.8
Llama (CT) 2025.06		93.3	90.5	96	94	79.9	86	95.3	90.7
gpt-3.5 (GPb+CT son) 2025.06		92.7	56.7	98.7	92.3	97.8	83.3	94.7	88
Haiku (GPb son) 2025.06		92.7	73.3	99.3	88.5	97.6	96.3	100	92.5
Haiku (GPb ol) 2025.06		92.7	79.3	100	88.8	88	88.7	87.3	89.3
Llama (Base) 2025.06		92.7	86.7	88	91.9	75.6	83.7	92	87.2
gpt-3.5 (base) 2025.06		92	68.7	84.7	84.1	58.9	74.3	90	79
Haiku (base) 2025.06		91.3	84.7	90	90.1	63.6	88.3	98.7	86.7
Haiku (CT) 2025.06		90.7	84.7	90.7	94	73.8	91	100	89.3
gpt-3.5 (GPb ol) 2025.06		89.3	77.3	100	84.5	86.4	69	94.7	85.9
gpt-3.5 (CT) 2025.06		88.7	77.3	68	87.3	64.7	69.3	90	77.9
gpt-3.5 (GPb son) 2025.06		88.7	62	100	84.8	85.8	78	98.7	85.4
Llama (GPb) 2025.06		87.3	87.3	100	89.5	91.1	88	87.3	90.1