Back to Models
deepseek logo
deepseek/deepseek-r1-0528
Not Available

R1 0528

May 28th update to the [original DeepSeek R1](/deepseek/deepseek-r1) Performance on par with [OpenAI o1](/openai/o1), but open-sourced and with fully open reasoning tokens. It's 671B parameters in size, with 37B active in an inference pass. Fully open-source model.

5/28/2025
163,840 tokens
#72 Text (Japanese)
Specifications

Modalities

Input
text
Output
text

Supported Parameters

frequency_penalty
include_reasoning
logit_bias
logprobs
max_tokens
min_p
presence_penalty
reasoning
repetition_penalty
response_format
seed
stop
structured_outputs
temperature
tool_choice
tools
top_k
top_logprobs
top_p

Max Output Tokens

65,536
Leaderboard
Text
๐Ÿ†OverallELO: 1,422
#104
๐Ÿ‡ฏ๐Ÿ‡ตJapaneseELO: 1,384
#72
๐Ÿ‡จ๐Ÿ‡ณChineseELO: 1,446
#117
๐Ÿ‡ฐ๐Ÿ‡ทKoreanELO: 1,349
#109
๐Ÿ‡ฌ๐Ÿ‡งEnglishELO: 1,432
#106
frenchELO: 1,423
#117
germanELO: 1,384
#116
spanishELO: 1,396
#116
russianELO: 1,415
#98
๐Ÿ’ปCodingELO: 1,464
#106
๐ŸงฎMathELO: 1,397
#140
โœ๏ธCreative WritingELO: 1,394
#99
๐Ÿ“Instruction FollowingELO: 1,392
#132
๐ŸŒถ๏ธHard PromptsELO: 1,433
#113
๐Ÿ’ฌMulti-TurnELO: 1,406
#130