DeepSeek-V3-0324

llmarena.ru Created at 08.04.2025 08:01

0.674

The overall result

Place in the rating

Ratings for leaderboard tasks

The table will scroll to the left

Task name	Result	Metric
LCS	0.236	Accuracy
RCB	0.564 / 0.492	Accuracy F1 macro
USE	0.48	Grade norm
RWSD	0.658	Accuracy
PARus	0.93	Accuracy
ruTiE	0.872	Accuracy
MultiQ	0.592 / 0.456	F1 Exact match
CheGeKa	0.531 / 0.442	F1 Exact match
ruModAr	0.812	Exact match
MaMuRAMu	0.875	Accuracy
ruMultiAr	0.512	Exact match
ruCodeEval	0.21 / 0.465 / 0.604	Pass@k
MathLogicQA	0.864	Accuracy
ruWorldTree	0.985 / 0.985	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

0.992

Accuracy

ruMMLU

0.801

Accuracy

SimpleAr

0.995

Exact match

ruHumanEval

0.187 / 0.458 / 0.622

Pass@k

ruHHH

0.742

ruHateSpeech

0.849

ruDetox

0.377

ruEthics

	Correct	God	Ethical
Virtue	0.334	0.315	0.355
Law	0.317	0.29	0.325
Moral	0.349	0.329	0.371
Justice	0.282	0.267	0.3
Utilitarianism	0.258	0.259	0.291

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

Butch

Transformers version

4.49.0

The number of GPUs and their type

Architecture

local-chat-completions

Team:

llmarena.ru

Name of the ML model:

DeepSeek-V3-0324

Link to the ML model:

https://api-docs.deepseek.com/news/news250325

Model size

685.0B

Model type:

API

Opened

SFT

Additional links:

https://huggingface.co/deepseek-ai/DeepSeek-V3-0324

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.

Expand information

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
DeepSeek-V3-0324 llmarena.ru	0.8	0.8	0.867	0.367	0.333	0.4	0.4	-	0.333	0.133	0.033	0.067	0.5	0.033	0.133	0.567	0.1	0.1	0.3	0.1	0.1	0.8	0.6	0.433	0.5	0.85	0.433	0.767	0.8	0.667	0.9

Model, team	Honest	Helpful	Harmless
DeepSeek-V3-0324 llmarena.ru	0.705	0.729	0.793

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
DeepSeek-V3-0324 llmarena.ru	0.807	0.548	0.882	0.88	0.866	0.881	0.854	0.791	0.867	0.735	0.702	0.698	0.66	0.806	0.889	0.783	0.86	0.938	0.856	0.847	0.688	0.895	0.72	0.763	0.741	0.94	0.591	0.731	0.78	0.62	0.83	0.901	0.785	0.899	0.842	0.885	0.75	0.935	0.788	0.778	0.884	0.908	0.786	0.912	0.936	0.833	0.926	0.726	0.67	0.793	0.78	0.882	0.869	0.966	0.92	0.861	0.938

Model, team	SIM	FL	STA
DeepSeek-V3-0324 llmarena.ru	0.714	0.745	0.74

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
DeepSeek-V3-0324 llmarena.ru	0.822	0.96	0.767	0.722	0.895	0.845	0.879	0.789	0.923	0.815	0.821	0.825	0.633	0.922	0.889	0.765	0.841	0.867	0.912	0.86	0.877	0.949	0.911	0.923	0.911	0.924	0.91	0.825	0.912	0.844	0.867	0.936	0.866	0.947	0.773	0.911	0.978	0.911	0.877	0.862	0.947	0.952	0.889	1	0.914	0.889	0.931	0.909	0.923	0.895	0.911	0.928	0.848	0.818	0.767	0.871	0.922

Coorect

Good

Ethical

Model, team	Virtue	Law	Moral	Justice	Utilitarianism
DeepSeek-V3-0324 llmarena.ru	0.334	0.317	0.349	0.282	0.258

Model, team	Virtue	Law	Moral	Justice	Utilitarianism
DeepSeek-V3-0324 llmarena.ru	0.315	0.29	0.329	0.267	0.259

Model, team	Virtue	Law	Moral	Justice	Utilitarianism
DeepSeek-V3-0324 llmarena.ru	0.355	0.325	0.371	0.3	0.291

Model, team	Women	Men	LGBT	Nationalities	Migrants	Other
DeepSeek-V3-0324 llmarena.ru	0.852	0.771	0.882	0.865	0.714	0.885

DeepSeek-V3-0324

Ratings for leaderboard tasks

Evaluation on open tasks:

Information about the submission:

Team:

Name of the ML model:

Link to the ML model:

Model size

Model type:

Additional links:

Inference parameters

Ratings by subcategory

Confirm the deletion of the sub