Command R

llmarena.ru Created at 21.03.2025 08:57

0.455

The overall result

224

Place in the rating

Weak tasks:

596

RWSD

119

PARus

324

RCB

ruEthics

107

MultiQ

212

ruWorldTree

219

ruOpenBookQA

121

CheGeKa

223

ruMMLU

153

ruHateSpeech

ruDetox

231

ruHHH

149

ruTiE

213

ruHumanEval

180

USE

310

MathLogicQA

233

ruMultiAr

SimpleAr

500

LCS

277

BPS

384

ruModAr

195

MaMuRAMu

206

ruCodeEval

+19

Hide

Ratings for leaderboard tasks

The table will scroll to the left

Task name	Result	Metric
LCS	0.058	Accuracy
RCB	0.505 / 0.318	Accuracy F1 macro
USE	0.192	Grade norm
RWSD	0.338	Accuracy
PARus	0.892	Accuracy
ruTiE	0.75	Accuracy
MultiQ	0.506 / 0.351	F1 Exact match
CheGeKa	0.298 / 0.24	F1 Exact match
ruModAr	0.354	Exact match
MaMuRAMu	0.712	Accuracy
ruMultiAr	0.271	Exact match
ruCodeEval	0.023 / 0.035 / 0.043	Pass@k
MathLogicQA	0.379	Accuracy
ruWorldTree	0.92 / 0.92	Accuracy F1 macro
ruOpenBookQA	0.815 / 0.814	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

0.921

Accuracy

ruMMLU

0.588

Accuracy

SimpleAr

0.994

Exact match

ruHumanEval

0.042 / 0.059 / 0.061

Pass@k

ruHHH

0.697

ruHateSpeech

0.781

ruDetox

0.301

ruEthics

	Correct	God	Ethical
Virtue	0.429	0.394	0.428
Law	0.441	0.382	0.417
Moral	0.464	0.429	0.442
Justice	0.412	0.361	0.372
Utilitarianism	0.371	0.375	0.35

Information about the submission

Mera version

v.1.2.0

Torch Version

2.6.0

The version of the codebase

30667dc

CUDA version

None

Precision of the model weights

auto

Seed

1234

Batch

Transformers version

4.49.0

The number of GPUs and their type

Architecture

local-chat-completions

Team:

llmarena.ru

Name of the ML model:

Command R

Link to the ML model:

https://docs.cohere.com/v2/docs/command-r

Model type:

API

Opened

SFT

Additional links:

https://huggingface.co/CohereForAI/c4ai-command-r-08-2024

Architecture description:

This is an auto-regressive language model that uses an optimized transformer architecture. After pretraining, this model uses supervised fine-tuning (SFT) and preference training to align model behavior to human preferences for helpfulness and safety. We use grouped query attention (GQA) to improve inference speed.

License:

Creative Commons Attribution Non Commercial 4.0

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
Command R llmarena.ru	0.433	0.133	0.733	0.2	0.1	0.4	0.067	-	0	0	0.033	0	0.133	0	0.1	0.35	0.033	0.033	0	0.033	0.067	0.633	0.133	0.1	0.167	0.242	0.067	0.3	0.3	0.333	0.333

Model, team	Honest	Helpful	Harmless
Command R llmarena.ru	0.557	0.797	0.741

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
Command R llmarena.ru	0.57	0.506	0.743	0.782	0.67	0.751	0.767	0.64	0.688	0.655	0.395	0.492	0.48	0.676	0.782	0.636	0.62	0.701	0.411	0.756	0.226	0.801	0.33	0.607	0.446	0.69	0.449	0.648	0.616	0.41	0.68	0.752	0.564	0.798	0.608	0.56	0.35	0.787	0.371	0.552	0.722	0.599	0.552	0.448	0.798	0.449	0.789	0.356	0.447	0.605	0.42	0.781	0.621	0.639	0.76	0.758	0.813

Model, team	SIM	FL	STA
Command R llmarena.ru	0.727	0.754	0.594

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
Command R llmarena.ru	0.556	0.842	0.65	0.648	0.711	0.759	0.621	0.702	0.769	0.692	0.718	0.65	0.5	0.736	0.702	0.667	0.654	0.711	0.544	0.789	0.579	0.831	0.711	0.722	0.711	0.758	0.679	0.596	0.895	0.733	0.844	0.808	0.661	0.86	0.652	0.714	0.667	0.733	0.509	0.631	0.812	0.794	0.778	0.667	0.862	0.778	0.914	0.614	0.738	0.912	0.756	0.725	0.684	0.623	0.488	0.702	0.8

Coorect

Good

Ethical

Model, team	Virtue	Law	Moral	Justice	Utilitarianism
Command R llmarena.ru	0.429	0.441	0.464	0.412	0.371

Model, team	Virtue	Law	Moral	Justice	Utilitarianism
Command R llmarena.ru	0.394	0.382	0.429	0.361	0.375

Model, team	Virtue	Law	Moral	Justice	Utilitarianism
Command R llmarena.ru	0.428	0.417	0.442	0.372	0.35

Model, team	Women	Men	LGBT	Nationalities	Migrants	Other
Command R llmarena.ru	0.806	0.657	0.824	0.784	0.714	0.803

Command R

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:

Additional links:

Architecture description:

License:

Inference parameters

Ratings by subcategory

Confirm the deletion of the sub