Multilingual Language Proficiency on INCLUDE base-44 benchmark leaderboard · SOTA2 Research

Multilingual Language Proficiency on INCLUDE base-44

64.8Average ScoreBielik-11B-v3-Instruct

Evaluation Results

Method
Bielik-11B-v3-Instruct2026.03	64.8	69
Bielik-11B-v3.0-Instruct2026.04	64.8	69
Qwen2.5-14B-Instruct2026.03	61.7	58.9
Qwen2.5-14B-Instruct2026.04	61.7	58.9
Bielik-11B-v3.02026.03	60.6	63.9
Bielik-11B-v3-Base-202507302026.04	60.6	63.9
phi-42026.03	58.8	49.6
phi-42026.04	58.8	49.6
Apertus-8B-Instruct-25092026.03	57.9	49.6
Apertus-8B-Instruct-25092026.04	57.9	49.6
Bielik-Minitron-7B-v3.0-Instruct2026.03	57.4	59.3
Bielik-Minitron-7B-v3.0-Instruct2026.04	57.4	59.3
Llama-3.1-8B-Instruct2026.03	55.3	53.8
Llama-3.1-8B-Instruct2026.04	55.3	53.8
EuroLLM-9B-Instruct2026.03	55.1	52
EuroLLM-9B-Instruct2026.04	55.1	52
Qwen2.5-7B-Instruct2026.03	54.4	52.2
Qwen2.5-7B-Instruct2026.04	54.4	52.2
Bielik-PL-11B-v3.0-Instruct2026.04	53.92	64.23
Mistral-Nemo-Instruct-24072026.03	53.2	48.4
Mistral-Nemo-Instruct-24072026.04	53.2	48.4
Bielik-11B-v2.6-Instruct2026.03	51.5	59.3
Bielik-11B-v2.6-Instruct2026.04	51.5	59.3
Mistral-Nemo-Base-24072026.03	51.2	44.9
Mistral-Nemo-Base-24072026.04	51.2	44.9
Bielik-PL-Minitron-7B-v3.0-Instruct2026.04	49.81	59.49
EuroLLM-9B2026.03	49.2	45.6
EuroLLM-9B2026.04	49.2	45.6
aya-expanse-8b2026.03	45.3	46.4
Mistral-7B-Instruct-v0.22026.03	45.3	44.7
aya-expanse-8b2026.04	45.3	46.4
Mistral-7B-Instruct-v0.22026.04	45.3	44.7
Bielik-11B-v22026.03	44.8	53.5
Bielik-11B-v22026.04	44.8	53.5
pllum-12b-nc-chat-2507152026.03	44.2	60.6
pllum-12b-nc-chat-2507152026.04	44.2	60.6
Mistral-7B-v0.22026.03	41.8	37.2
Mistral-7B-v0.22026.04	41.8	37.2
pllum-12b-nc-base-2507152026.03	37.8	52.7
pllum-12b-nc-base-2507152026.04	37.8	52.7
Bielik-4.5B-v3.02026.03	35.9	48.7
Bielik-4.5B-v32026.04	35.9	48.7
PLLuM-12B-base-2508012026.03	35.5	44.5
PLLuM-12B-base-2508012026.04	35.5	44.5
Llama-PLLuM-8B-base-2508012026.03	30	37.2
Llama-PLLuM-8B-base-2508012026.04	30	37.2