DeepSeek

R1 0528

by DeepSeektext->text4 endpoints
Neura Intelligence Index
47.1/ 100
Rank
#176
Confidence
medium
Benchmarks
7(18% coverage)

Overview

May 28th update to the original DeepSeek R1 Performance on par with OpenAI o1, but open-sourced and with fully open reasoning tokens. It's 671B parameters in size, with 37B active...

Capabilities

Text generation
Tool use / Function calling
Structured output
Extended reasoning
Prompt caching
Long context
Top-P sampling
Stop sequences
Deterministic seed
Log probabilities

Modalities

Input
Text
Output
Text

Technical Specifications

Context Window
163.8K tokens
Max Output
32.8K tokens
Knowledge Cutoff
2025-03-31
Tokenizer
DeepSeek
Uptime (24h)
9992.1%
Parameters
671000000000B
License
mitOpen Weights
Country
🇨🇳CN
Release Date
May 28, 20251y

Supported Parameters

frequency_penaltyinclude_reasoninglogit_biaslogprobsmax_tokensmin_ppresence_penaltyreasoningrepetition_penaltyresponse_formatseedstopstructured_outputstemperaturetool_choicetoolstop_ktop_logprobstop_p

Benchmark Performance

general

BenchmarkScoreSource
SimpleQA
92.3
verified

coding

BenchmarkScoreSource
SWE-bench Verified
44.6
verified
TerminalBench
5.7
verified

math

BenchmarkScoreSource
AIME 2025
87.5
verified

reasoning

BenchmarkScoreSource
GPQA Diamond
81
verified
Humanity's Last Exam
17.7
verified

agent

BenchmarkScoreSource
BrowseComp
8.9
verified

Performance Over Time

Pricing

Per 1M tokens
Input$0.50
Output$2.15
Cache read$0.35
Added
May 28, 2025
Last synced 8/27/2026