GigaChat 2 Max

GIGACHAT Created at 12.03.2025 11:33

0.67

The overall result

Place in the rating

Ratings for leaderboard tasks

The table will scroll to the left

Task name	Result	Metric
LCS	0.15	Accuracy
RCB	0.575 / 0.403	Accuracy F1 macro
USE	0.581	Grade norm
RWSD	0.642	Accuracy
PARus	0.93	Accuracy
ruTiE	0.901	Accuracy
MultiQ	0.534 / 0.4	F1 Exact match
CheGeKa	0.414 / 0.344	F1 Exact match
ruModAr	0.963	Exact match
MaMuRAMu	0.864	Accuracy
ruMultiAr	0.476	Exact match
ruCodeEval	0.505 / 0.559 / 0.585	Pass@k
MathLogicQA	0.815	Accuracy
ruWorldTree	0.996 / 0.996	Accuracy F1 macro
ruOpenBookQA	0.948 / 0.761	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

0.993

Accuracy

ruMMLU

0.795

Accuracy

SimpleAr

0.994

Exact match

ruHumanEval

0.502 / 0.55 / 0.579

Pass@k

ruHHH

0.86

ruHateSpeech

0.645

ruDetox

0.167

ruEthics

	Correct	God	Ethical
Virtue	0.356	0.349	0.437
Law	0.374	0.346	0.415
Moral	0.395	0.366	0.457
Justice	0.326	0.316	0.388
Utilitarianism	0.311	0.319	0.362

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

Seed

1234

Butch

Transformers version

4.49.0

The number of GPUs and their type

Architecture

gigachat_llms

Team:

GIGACHAT

Name of the ML model:

GigaChat 2 Max

Link to the ML model:

https://developers.sber.ru/docs/ru/gigachat/models

Model type:

Closed

API

SFT

Architecture description:

GigaChat 2 Max is a new Large Language Model (LLM) by Sber. The model will be available for B2C users https://giga.chat/ and B2B users via API soon https://developers.sber.ru/docs/ru/gigachat/api/tariffs.

Description of the training:

Pretrain data:

License:

Proprietary model by Sber

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.

Description of the template:
API Default

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
GigaChat 2 Max GIGACHAT	0.9	0.567	0.767	0.5	0.667	0.667	0.533	-	0.5	0.2	0.367	0.267	0.667	0.467	0.233	0.817	0.4	0.4	0.767	0.267	0.333	0.667	0.567	0.3	0.233	0.775	0.6	0.833	0.8	0.7	0.867

Model, team	Honest	Helpful	Harmless
GigaChat 2 Max GIGACHAT	0.803	0.831	0.948

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
GigaChat 2 Max GIGACHAT	0.733	0.542	0.921	0.902	0.889	0.871	0.854	0.839	0.895	0.794	0.737	0.698	0.59	0.806	0.898	0.812	0.78	0.938	0.7	0.832	0.655	0.871	0.67	0.827	0.768	0.92	0.604	0.731	0.751	0.61	0.86	0.86	0.828	0.899	0.875	0.897	0.7	0.929	0.762	0.788	0.909	0.857	0.752	0.867	0.919	0.782	0.853	0.689	0.631	0.81	0.79	0.831	0.872	0.929	0.95	0.83	0.922

Model, team	SIM	FL	STA
GigaChat 2 Max GIGACHAT	0.277	0.725	0.764

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
GigaChat 2 Max GIGACHAT	0.867	0.95	0.85	0.713	0.882	0.879	0.793	0.754	0.904	0.877	0.846	0.817	0.592	0.93	0.86	0.778	0.804	0.822	0.825	0.86	0.895	0.949	0.889	0.899	0.978	0.939	0.897	0.825	0.895	0.956	0.911	0.885	0.875	0.912	0.788	0.964	0.889	0.889	0.754	0.831	0.902	0.889	0.867	1	0.879	0.889	0.914	0.955	0.938	0.947	0.822	0.913	0.861	0.792	0.651	0.877	0.922

Coorect

Good

Ethical

Model, team	Virtue	Law	Moral	Justice	Utilitarianism
GigaChat 2 Max GIGACHAT	0.356	0.374	0.395	0.326	0.311

Model, team	Virtue	Law	Moral	Justice	Utilitarianism
GigaChat 2 Max GIGACHAT	0.349	0.346	0.366	0.316	0.319

Model, team	Virtue	Law	Moral	Justice	Utilitarianism
GigaChat 2 Max GIGACHAT	0.437	0.415	0.457	0.388	0.362

Model, team	Women	Men	LGBT	Nationalities	Migrants	Other
GigaChat 2 Max GIGACHAT	0.852	0.629	0	0.162	0.286	0.803

GigaChat 2 Max

Ratings for leaderboard tasks

Evaluation on open tasks:

Information about the submission:

Team:

Name of the ML model:

Link to the ML model:

Model type:

Architecture description:

Description of the training:

Pretrain data:

License:

Inference parameters

Ratings by subcategory

Confirm the deletion of the sub