Qwen3-30B-A3B-Instruct-2507

MERA Created at 12.02.2026 00:16
0.492
The overall result
160
Place in the rating
Weak tasks:
484
RWSD
77
PARus
111
RCB
279
ruEthics
117
MultiQ
38
ruWorldTree
61
ruOpenBookQA
164
CheGeKa
101
ruMMLU
125
ruHateSpeech
174
ruDetox
175
ruHHH
46
ruTiE
108
USE
126
MathLogicQA
350
ruMultiAr
395
SimpleAr
415
LCS
408
BPS
402
ruModAr
100
MaMuRAMu
+17
Hide

Ratings for leaderboard tasks

The table will scroll to the left

Task name Result Metric
LCS 0.084 Accuracy
RCB 0.566 / 0.505 Accuracy F1 macro
USE 0.292 Grade norm
RWSD 0.481 Accuracy
PARus 0.916 Accuracy
ruTiE 0.867 Accuracy
MultiQ 0.493 / 0.296 F1 Exact match
CheGeKa 0.189 / 0.139 F1 Exact match
ruModAr 0.326 Exact match
MaMuRAMu 0.801 Accuracy
ruMultiAr 0.203 Exact match
ruCodeEval 0 / 0 / 0 Pass@k
MathLogicQA 0.506 Accuracy
ruWorldTree 0.989 / 0.792 Accuracy F1 macro
ruOpenBookQA 0.925 / 0.925 Accuracy F1 macro

Evaluation on open tasks:

Go to the ratings by subcategory

The table will scroll to the left

Task name Result Metric
BPS 0.769 Accuracy
ruMMLU 0.71 Accuracy
SimpleAr 0.851 Exact match
ruHumanEval 0 / 0 / 0 Pass@k
ruHHH 0.753
ruHateSpeech 0.796
ruDetox 0.22
ruEthics
Correct God Ethical
Virtue 0.314 0.327 0.245
Law 0.316 0.306 0.253
Moral 0.357 0.344 0.282
Justice 0.286 0.28 0.205
Utilitarianism 0.292 0.294 0.224

Information about the submission

Mera version
v1.2.0
Torch Version
2.8
The version of the codebase
7030e16
CUDA version
12.8
Precision of the model weights
bfloat16
Seed
1234
Batch
1
Transformers version
4.56.1
The number of GPUs and their type
4 x NVIDIA A100 80GB
Architecture
local-chat-completions

Team:

MERA

Name of the ML model:

Qwen3-30B-A3B-Instruct-2507

Model size

30.0B

Model type:

Opened

SFT

MoE

License:

apache-2.0

Inference parameters

Generation Parameters:
simplear - do_sample=false;until=["\n"]; \nchegeka - do_sample=false;until=["\n"]; \nrudetox - do_sample=false;until=["\n"]; \nrumultiar - do_sample=false;until=["\n"]; \nuse - do_sample=false;until=["\n","."]; \nmultiq - do_sample=false;until=["\n"]; \nrumodar - do_sample=false;until=["\n"]; \nruhumaneval - do_sample=true;until=["\nclass","\ndef","\n#","\nif","\nprint"];temperature=0.6; \nrucodeeval - do_sample=true;until=["\nclass","\ndef","\n#","\nif","\nprint"];temperature=0.6;

System prompt:
Реши задачу по инструкции ниже. Не давай никаких объяснений и пояснений к своему ответу. Не пиши ничего лишнего. Пиши только то, что указано в инструкции. Если по инструкции нужно решить пример, то напиши только числовой ответ без хода решения и пояснений. Если по инструкции нужно вывести букву, цифру или слово, выведи только его. Если по инструкции нужно выбрать один из вариантов ответа и вывести букву или цифру, которая ему соответствует, то выведи только эту букву или цифру, не давай никаких пояснений, не добавляй знаки препинания, только 1 символ в ответе. Если по инструкции нужно дописать код функции на языке Python, пиши сразу код, соблюдая отступы так, будто ты продолжаешь функцию из инструкции, не давай пояснений, не пиши комментарии, используй только аргументы из сигнатуры функции в инструкции, не пробуй считывать данные через функцию input. Не извиняйся, не строй диалог. Выдавай только ответ и ничего больше.

Description of the template:
{%- if tools %} {{- '<|im_start|>system' }} {%- if messages[0].role == 'system' %} {{- messages[0].content + '' }} {%- endif %} {{- "# ToolsYou may call one or more functions to assist with the user query.You are provided with function signatures within <tools></tools> XML tags:<tools>" }} {%- for tool in tools %} {{- "" }} {{- tool | tojson }} {%- endfor %} {{- "</tools>For each function call, return a json object with function name and arguments within <tool_call></tool_call> XML tags:<tool_call>{"name": <function-name>, "arguments": <args-json-object>}</tool_call><|im_end|>" }} {%- else %} {%- if messages[0].role == 'system' %} {{- '<|im_start|>system' + messages[0].content + '<|im_end|>' }} {%- endif %} {%- endif %} {%- for message in messages %} {%- if message.content is string %} {%- set content = message.content %} {%- else %} {%- set content = '' %} {%- endif %} {%- if (message.role == "user") or (message.role == "system" and not loop.first) %} {{- '<|im_start|>' + message.role + '' + content + '<|im_end|>' + '' }} {%- elif message.role == "assistant" %} {{- '<|im_start|>' + message.role + '' + content }} {%- if message.tool_calls %} {%- for tool_call in message.tool_calls %} {%- if (loop.first and content) or (not loop.first) %} {{- '' }} {%- endif %} {%- if tool_call.function %} {%- set tool_call = tool_call.function %} {%- endif %} {{- '<tool_call>{"name": "' }} {{- tool_call.name }} {{- '", "arguments": ' }} {%- if tool_call.arguments is string %} {{- tool_call.arguments }} {%- else %} {{- tool_call.arguments | tojson }} {%- endif %} {{- '}</tool_call>' }} {%- endfor %} {%- endif %} {{- '<|im_end|>' }} {%- elif message.role == "tool" %} {%- if loop.first or (messages[loop.index0 - 1].role != "tool") %} {{- '<|im_start|>user' }} {%- endif %} {{- '<tool_response>' }} {{- content }} {{- '</tool_response>' }} {%- if loop.last or (messages[loop.index0 + 1].role != "tool") %} {{- '<|im_end|>' }} {%- endif %} {%- endif %} {%- endfor %} {%- if add_generation_prompt %} {{- '<|im_start|>assistant' }} {%- endif %}

Ratings by subcategory

Metric: Grade Norm
Model, team 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 8_0 8_1 8_2 8_3 8_4
Qwen3-30B-A3B-Instruct-2507
MERA
0.533 0.4 0.867 0.167 0.1 0.367 0.133 - 0.033 0 0 0 0.267 0.133 0.067 0.517 0.033 0.033 0 0.033 0.067 0.767 0.433 0.233 0 0.633 0.2 0.4 0.5 0.267 0.333
Model, team Honest Helpful Harmless
Qwen3-30B-A3B-Instruct-2507
MERA
0.689 0.847 0.724
Model, team Anatomy Virology Astronomy Marketing Nutrition Sociology Management Philosophy Prehistory Human aging Econometrics Formal logic Global facts Jurisprudence Miscellaneous Moral disputes Business ethics Biology (college) Physics (college) Human Sexuality Moral scenarios World religions Abstract algebra Medicine (college) Machine learning Medical genetics Professional law PR Security studies Chemistry (школьная) Computer security International law Logical fallacies Politics Clinical knowledge Conceptual_physics Math (college) Biology (high school) Physics (high school) Chemistry (high school) Geography (high school) Professional medicine Electrical engineering Elementary mathematics Psychology (high school) Statistics (high school) History (high school) Math (high school) Professional accounting Professional psychology Computer science (college) World history (high school) Macroeconomics Microeconomics Computer science (high school) European history Government and politics
Qwen3-30B-A3B-Instruct-2507
MERA
0.748 0.542 0.875 0.88 0.82 0.851 0.825 0.768 0.821 0.709 0.658 0.508 0.48 0.778 0.799 0.74 0.74 0.903 0.411 0.824 0.493 0.795 0.15 0.757 0.616 0.87 0.508 0.676 0.792 0.48 0.78 0.835 0.712 0.848 0.785 0.778 0.2 0.887 0.629 0.675 0.854 0.853 0.697 0.684 0.906 0.685 0.897 0.43 0.564 0.749 0.65 0.865 0.795 0.912 0.66 0.824 0.865
Model, team SIM FL STA
Qwen3-30B-A3B-Instruct-2507
MERA
0.471 0.72 0.702
Model, team Anatomy Virology Astronomy Marketing Nutrition Sociology Managment Philosophy Pre-History Gerontology Econometrics Formal logic Global facts Jurisprudence Miscellaneous Moral disputes Business ethics Bilology (college) Physics (college) Human sexuality Moral scenarios World religions Abstract algebra Medicine (college) Machine Learning Genetics Professional law PR Security Chemistry (college) Computer security International law Logical fallacies Politics Clinical knowledge Conceptual physics Math (college) Biology (high school) Physics (high school) Chemistry (high school) Geography (high school) Professional medicine Electrical Engineering Elementary mathematics Psychology (high school) Statistics (high school) History (high school) Math (high school) Professional Accounting Professional psychology Computer science (college) World history (high school) Macroeconomics Microeconomics Computer science (high school) Europe History Government and politics
Qwen3-30B-A3B-Instruct-2507
MERA
0.622 0.881 0.717 0.722 0.934 0.845 0.741 0.754 0.808 0.785 0.782 0.808 0.608 0.814 0.743 0.728 0.766 0.822 0.561 0.86 0.789 0.847 0.844 0.876 0.756 0.894 0.821 0.684 0.912 0.911 0.844 0.872 0.821 0.912 0.682 0.875 0.889 0.844 0.684 0.769 0.865 0.921 0.889 0.556 0.879 0.844 0.914 0.818 0.815 0.93 0.889 0.841 0.861 0.753 0.442 0.713 0.867
Coorect
Good
Ethical
Model, team Virtue Law Moral Justice Utilitarianism
Qwen3-30B-A3B-Instruct-2507
MERA
0.314 0.316 0.357 0.286 0.292
Model, team Virtue Law Moral Justice Utilitarianism
Qwen3-30B-A3B-Instruct-2507
MERA
0.327 0.306 0.344 0.28 0.294
Model, team Virtue Law Moral Justice Utilitarianism
Qwen3-30B-A3B-Instruct-2507
MERA
0.245 0.253 0.282 0.205 0.224
Model, team Women Men LGBT Nationalities Migrants Other
Qwen3-30B-A3B-Instruct-2507
MERA
0.806 0.743 0.706 0.784 0.714 0.852