Mistral Small 3.1 24B Instruct

llmarena.ru Created at 21.03.2025 08:59
0.526
The overall result
120
Place in the rating
Weak tasks:
250
RWSD
83
PARus
124
RCB
257
ruEthics
73
MultiQ
101
ruWorldTree
98
ruOpenBookQA
106
CheGeKa
98
ruMMLU
173
ruHateSpeech
23
ruDetox
130
ruHHH
60
ruTiE
281
ruHumanEval
64
USE
116
MathLogicQA
156
ruMultiAr
80
SimpleAr
177
LCS
323
BPS
431
ruModAr
84
MaMuRAMu
260
ruCodeEval
+19
Hide

Ratings for leaderboard tasks

The table will scroll to the left

Task name Result Metric
LCS 0.13 Accuracy
RCB 0.564 / 0.52 Accuracy F1 macro
USE 0.354 Grade norm
RWSD 0.542 Accuracy
PARus 0.912 Accuracy
ruTiE 0.861 Accuracy
MultiQ 0.569 / 0.416 F1 Exact match
CheGeKa 0.324 / 0.25 F1 Exact match
ruModAr 0.249 Exact match
MaMuRAMu 0.813 Accuracy
ruMultiAr 0.315 Exact match
ruCodeEval 0.002 / 0.008 / 0.012 Pass@k
MathLogicQA 0.519 Accuracy
ruWorldTree 0.968 / 0.968 Accuracy F1 macro
ruOpenBookQA 0.903 / 0.902 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 0.894 Accuracy
ruMMLU 0.714 Accuracy
SimpleAr 0.993 Exact match
ruHumanEval 0.007 / 0.011 / 0.012 Pass@k
ruHHH 0.803
ruHateSpeech 0.774
ruDetox 0.356
ruEthics
Correct God Ethical
Virtue 0.283 0.269 0.352
Law 0.27 0.248 0.326
Moral 0.296 0.28 0.367
Justice 0.248 0.239 0.313
Utilitarianism 0.215 0.244 0.3

Information about the submission

Mera version
v.1.2.0
Torch Version
2.5.1
The version of the codebase
30667dc
CUDA version
12.4
Precision of the model weights
auto
Seed
1234
Batch
1
Transformers version
4.49.0
The number of GPUs and their type
0
Architecture
local-chat-completions

Team:

llmarena.ru

Name of the ML model:

Mistral Small 3.1 24B Instruct

Model size

24.0M

Model type:

API

Opened

SFT

Additional links:

https://huggingface.co/mistralai/Mistral-Small-3.1-24B-Instruct-2503

Inference parameters

Generation Parameters:
rucodeeval - do_sample=true;temperature=0.6;until=["\nclass","\ndef","\n#","\nif","\nprint"]; \nruhumaneval - do_sample=true;temperature=0.6;until=["\nclass","\ndef","\n#","\nif","\nprint"];

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.

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
Mistral Small 3.1 24B Instruct
llmarena.ru
0.4 0.467 0.733 0.233 0.167 0.5 0.133 - 0.2 0 0 0 0.333 0.133 0.1 0.417 0.033 0.067 0 0.033 0.1 0.8 0.467 0.367 0.333 0.742 0.333 0.5 0.667 0.533 0.6
Model, team Honest Helpful Harmless
Mistral Small 3.1 24B Instruct
llmarena.ru
0.787 0.746 0.879
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
Mistral Small 3.1 24B Instruct
llmarena.ru
0.667 0.488 0.868 0.842 0.807 0.826 0.816 0.73 0.784 0.7 0.614 0.524 0.59 0.759 0.849 0.705 0.76 0.882 0.644 0.809 0.569 0.865 0.43 0.728 0.598 0.8 0.514 0.657 0.731 0.53 0.72 0.81 0.73 0.838 0.785 0.778 0.49 0.881 0.616 0.729 0.818 0.824 0.648 0.716 0.886 0.671 0.887 0.463 0.557 0.715 0.6 0.857 0.81 0.828 0.85 0.806 0.865
Model, team SIM FL STA
Mistral Small 3.1 24B Instruct
llmarena.ru
0.757 0.681 0.721
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
Mistral Small 3.1 24B Instruct
llmarena.ru
0.6 0.901 0.75 0.685 0.842 0.724 0.672 0.772 0.904 0.769 0.833 0.8 0.525 0.891 0.842 0.691 0.766 0.867 0.737 0.895 0.719 0.898 0.889 0.876 0.867 0.818 0.885 0.807 0.895 0.8 0.8 0.872 0.804 0.947 0.682 0.857 0.911 0.8 0.667 0.769 0.878 0.921 0.822 0.956 0.948 0.889 0.879 0.909 0.877 0.965 0.911 0.855 0.785 0.675 0.581 0.754 0.856
Coorect
Good
Ethical
Model, team Virtue Law Moral Justice Utilitarianism
Mistral Small 3.1 24B Instruct
llmarena.ru
0.283 0.27 0.296 0.248 0.215
Model, team Virtue Law Moral Justice Utilitarianism
Mistral Small 3.1 24B Instruct
llmarena.ru
0.269 0.248 0.28 0.239 0.244
Model, team Virtue Law Moral Justice Utilitarianism
Mistral Small 3.1 24B Instruct
llmarena.ru
0.352 0.326 0.367 0.313 0.3
Model, team Women Men LGBT Nationalities Migrants Other
Mistral Small 3.1 24B Instruct
llmarena.ru
0.815 0.657 0.765 0.676 0.857 0.82