GeekRadar / The comparison desk

COMPARE.

AI models

deepseek

DeepSeek R1 0528

AI model Released May 28, 2025

Context window: 163,840 tokens
Maximum output: 32,768 tokens
Open weights: Yes

Capability & preference

Human preference · Arena

Overall text preferences from head-to-head votes. Not factual accuracy or a task-success rate.

DeepSeek R1 0528 1427.7 deepseek-r1-0528 · 95% interval 1422.1–1433.4 · 18,167 votes · 2026-10-15

Recorded results

Only the same benchmark, version, workload, scenario and unit share a row. Missing results are not zero. System results include the full submitted hardware and software configuration.

Arena text preference / chinesehuman preference · 2026-10-15 · rating

DeepSeek R1 0528

1,441.91 rating

Confidence interval: 1,424.29 – 1,459.53

1188 votes

Oct 15, 2026

Category: chinese
Evaluated configuration: deepseek-r1-0528
Evaluation type: human preference

Arena text preference / codinghuman preference · 2026-10-15 · rating

DeepSeek R1 0528

1,426.90 rating

Confidence interval: 1,415.49 – 1,438.30

2690 votes

Oct 15, 2026

Category: coding
Evaluated configuration: deepseek-r1-0528
Evaluation type: human preference

Arena text preference / creative_writinghuman preference · 2026-10-15 · rating

DeepSeek R1 0528

1,405.39 rating

Confidence interval: 1,392.97 – 1,417.81

2408 votes

Oct 15, 2026

Category: creative_writing
Evaluated configuration: deepseek-r1-0528
Evaluation type: human preference

Arena text preference / englishhuman preference · 2026-10-15 · rating

DeepSeek R1 0528

1,437.14 rating

Confidence interval: 1,429.79 – 1,444.49

8163 votes

Oct 15, 2026

Category: english
Evaluated configuration: deepseek-r1-0528
Evaluation type: human preference

Arena text preference / exclude_tieshuman preference · 2026-10-15 · rating

DeepSeek R1 0528

1,419.87 rating

Confidence interval: 1,412.06 – 1,427.67

13077 votes

Oct 15, 2026

Category: exclude_ties
Evaluated configuration: deepseek-r1-0528
Evaluation type: human preference

Arena text preference / experthuman preference · 2026-10-15 · rating

DeepSeek R1 0528

1,393.67 rating

Confidence interval: 1,374.01 – 1,413.32

886 votes

Oct 15, 2026

Category: expert
Evaluated configuration: deepseek-r1-0528
Evaluation type: human preference

Arena text preference / frenchhuman preference · 2026-10-15 · rating

DeepSeek R1 0528

1,417.01 rating

Confidence interval: 1,379.45 – 1,454.57

237 votes

Oct 15, 2026

Category: french
Evaluated configuration: deepseek-r1-0528
Evaluation type: human preference

Arena text preference / germanhuman preference · 2026-10-15 · rating

DeepSeek R1 0528

1,404.23 rating

Confidence interval: 1,379.45 – 1,429.02

597 votes

Oct 15, 2026

Category: german
Evaluated configuration: deepseek-r1-0528
Evaluation type: human preference

Arena text preference / hard_promptshuman preference · 2026-10-15 · rating

DeepSeek R1 0528

1,415.35 rating

Confidence interval: 1,407.67 – 1,423.04

6880 votes

Oct 15, 2026

Category: hard_prompts
Evaluated configuration: deepseek-r1-0528
Evaluation type: human preference

Arena text preference / hard_prompts_englishhuman preference · 2026-10-15 · rating

DeepSeek R1 0528

1,421.66 rating

Confidence interval: 1,410.89 – 1,432.43

3106 votes

Oct 15, 2026

Category: hard_prompts_english
Evaluated configuration: deepseek-r1-0528
Evaluation type: human preference

Arena text preference / industry_business_and_management_and_financial_operationshuman preference · 2026-10-15 · rating

DeepSeek R1 0528

1,401.54 rating

Confidence interval: 1,390.35 – 1,412.73

2852 votes

Oct 15, 2026

Category: industry_business_and_management_and_financial_operations
Evaluated configuration: deepseek-r1-0528
Evaluation type: human preference

Arena text preference / industry_entertainment_and_sports_and_mediahuman preference · 2026-10-15 · rating

DeepSeek R1 0528

1,400.39 rating

Confidence interval: 1,389.75 – 1,411.03

3182 votes

Oct 15, 2026

Category: industry_entertainment_and_sports_and_media
Evaluated configuration: deepseek-r1-0528
Evaluation type: human preference

Arena text preference / industry_legal_and_governmenthuman preference · 2026-10-15 · rating

DeepSeek R1 0528

1,418.38 rating

Confidence interval: 1,402.15 – 1,434.62

1370 votes

Oct 15, 2026

Category: industry_legal_and_government
Evaluated configuration: deepseek-r1-0528
Evaluation type: human preference

Arena text preference / industry_life_and_physical_and_social_sciencehuman preference · 2026-10-15 · rating

DeepSeek R1 0528

1,430.21 rating

Confidence interval: 1,419.64 – 1,440.77

3388 votes

Oct 15, 2026

Category: industry_life_and_physical_and_social_science
Evaluated configuration: deepseek-r1-0528
Evaluation type: human preference

Arena text preference / industry_mathematicalhuman preference · 2026-10-15 · rating

DeepSeek R1 0528

1,415.19 rating

Confidence interval: 1,395.89 – 1,434.50

919 votes

Oct 15, 2026

Category: industry_mathematical
Evaluated configuration: deepseek-r1-0528
Evaluation type: human preference

Arena text preference / industry_medicine_and_healthcarehuman preference · 2026-10-15 · rating

DeepSeek R1 0528

1,444.96 rating

Confidence interval: 1,427.45 – 1,462.47

1229 votes

Oct 15, 2026

Category: industry_medicine_and_healthcare
Evaluated configuration: deepseek-r1-0528
Evaluation type: human preference

Arena text preference / industry_software_and_it_serviceshuman preference · 2026-10-15 · rating

DeepSeek R1 0528

1,434.79 rating

Confidence interval: 1,426.44 – 1,443.14

5530 votes

Oct 15, 2026

Category: industry_software_and_it_services
Evaluated configuration: deepseek-r1-0528
Evaluation type: human preference

Arena text preference / industry_writing_and_literature_and_languagehuman preference · 2026-10-15 · rating

DeepSeek R1 0528

1,401.57 rating

Confidence interval: 1,392.08 – 1,411.06

4199 votes

Oct 15, 2026

Category: industry_writing_and_literature_and_language
Evaluated configuration: deepseek-r1-0528
Evaluation type: human preference

Arena text preference / instruction_followinghuman preference · 2026-10-15 · rating

DeepSeek R1 0528

1,381.59 rating

Confidence interval: 1,372.07 – 1,391.12

4015 votes

Oct 15, 2026

Category: instruction_following
Evaluated configuration: deepseek-r1-0528
Evaluation type: human preference

Arena text preference / japanesehuman preference · 2026-10-15 · rating

DeepSeek R1 0528

1,390.91 rating

Confidence interval: 1,366.06 – 1,415.75

620 votes

Oct 15, 2026

Category: japanese
Evaluated configuration: deepseek-r1-0528
Evaluation type: human preference

Arena text preference / koreanhuman preference · 2026-10-15 · rating

DeepSeek R1 0528

1,359.94 rating

Confidence interval: 1,325.50 – 1,394.38

323 votes

Oct 15, 2026

Category: korean
Evaluated configuration: deepseek-r1-0528
Evaluation type: human preference

Arena text preference / longer_queryhuman preference · 2026-10-15 · rating

DeepSeek R1 0528

1,391.27 rating

Confidence interval: 1,380.69 – 1,401.85

3129 votes

Oct 15, 2026

Category: longer_query
Evaluated configuration: deepseek-r1-0528
Evaluation type: human preference

Arena text preference / mathhuman preference · 2026-10-15 · rating

DeepSeek R1 0528

1,400.01 rating

Confidence interval: 1,380.08 – 1,419.95

861 votes

Oct 15, 2026

Category: math
Evaluated configuration: deepseek-r1-0528
Evaluation type: human preference

Arena text preference / multi_turnhuman preference · 2026-10-15 · rating

DeepSeek R1 0528

1,404.37 rating

Confidence interval: 1,393.40 – 1,415.34

2848 votes

Oct 15, 2026

Category: multi_turn
Evaluated configuration: deepseek-r1-0528
Evaluation type: human preference

Arena text preference / non_englishhuman preference · 2026-10-15 · rating

DeepSeek R1 0528

1,411.83 rating

Confidence interval: 1,405.00 – 1,418.66

9998 votes

Oct 15, 2026

Category: non_english
Evaluated configuration: deepseek-r1-0528
Evaluation type: human preference

Arena text preference / overallhuman preference · 2026-10-15 · rating

DeepSeek R1 0528

1,427.70 rating

Confidence interval: 1,422.05 – 1,433.36

18167 votes

Oct 15, 2026

Category: overall
Evaluated configuration: deepseek-r1-0528
Evaluation type: human preference

Arena text preference / polishhuman preference · 2026-10-15 · rating

DeepSeek R1 0528

1,428.20 rating

Confidence interval: 1,414.05 – 1,442.35

1980 votes

Oct 15, 2026

Category: polish
Evaluated configuration: deepseek-r1-0528
Evaluation type: human preference

Arena text preference / russianhuman preference · 2026-10-15 · rating

DeepSeek R1 0528

1,422.89 rating

Confidence interval: 1,408.23 – 1,437.55

1523 votes

Oct 15, 2026

Category: russian
Evaluated configuration: deepseek-r1-0528
Evaluation type: human preference

Arena text preference / spanishhuman preference · 2026-10-15 · rating

DeepSeek R1 0528

1,411.37 rating

Confidence interval: 1,375.87 – 1,446.87

261 votes

Oct 15, 2026

Category: spanish
Evaluated configuration: deepseek-r1-0528
Evaluation type: human preference

Run it locally

Inference · memory estimate

DeepSeek R1 0528

Not enough verified architecture information for a reliable VRAM estimate.

API cost

USD / month

Input tokens / monthOutput tokens / month

Standard text-token estimate. Excludes taxes, caching, batch discounts, reasoning and multimodal surcharges. Provider limits and regional availability may differ.

DeepSeek R1 0528

Provider offer

DeepSeek R1 0528 / Pricing tiers 1

Provider / scope	Rate	Currency	Amount / unit
DeepInfradeepseek-ai/DeepSeek-R1-0528Standard	Cached input	USD	0.35 / 1M tokens
	Input	USD	0.5 / 1M tokens
	Output	USD	2.15 / 1M tokens

Capabilities

Context window: 163,840 tokens
Maximum output: 32,768 tokens
Reasoning: Yes
Tool calling: Yes
Open weights: Yes
Model license: MIT

Input modalities

text

Output modalities

text

Processing & platform

Architecture: deepseek_v3