Compare / DeepSeek-R1 vs DeepSeek-R1-0528

DeepSeek-R1 vs DeepSeek-R1-0528

Which is cheaper, which is more capable, and which does better at each kind of work, from 16 results on 3 sources that measured both models.

DeepSeekDeepSeek-R1 Released 20 Jan 2025 · Not on sale
DeepSeekDeepSeek-R1-0528 Released 28 May 2025 · $0.50 in · $2.18 out
DeepSeek-R1 wins 35 too close to callDeepSeek-R1-0528 wins 8

Of the 16 results both models have, the better value on each. Results whose reported ranges overlap are too close to call.

Results as of 19 November 2026 (catalogue release 2026-11-19-3608133e8331).

DeepSeek-R1 or DeepSeek-R1-0528? The short answer

DeepSeek-R1-0528 is the newer version; they score about the same at about the same price.

  • DeepSeek-R1-0528 scores higher on the Artificial Analysis Intelligence Index: 13.1 against 11.4.
  • DeepSeek-R1-0528 is newer: released 28 May 2025, against 20 Jan 2025 for DeepSeek-R1.
  • On third-party benchmarks, on the results both models have: DeepSeek-R1-0528 does better for reasoning and knowledge.

Head to head

11.4
Intelligence Index
13.1

Every fact side by side

DeepSeek-R1DeepSeek-R1-0528
Price per million tokensNot on sale$0.50 in · $2.18 out
Blended (3 in : 1 out)–$0.92
Cheapest host, blended–$0.91 (DeepInfra, FP4)
Intelligence Index11.413.1
Spring Prompt overall––
Output speed, tokens a second––
Context window–163,840 tokens
Released20 Jan 202528 May 2025 (newer)
WeightsClosedOpen
Developer based inChinaChina

Prices are the developer's list price where we have read it, otherwise the typical host on OpenRouter; see every model's price. Intelligence and speed from Artificial Analysis.

Which is better for what

Each model's rank among the models on sale today, for each kind of work, from our own benchmarks and licensed sources. Further right is better; the better rank is in green.

Agents and tool useMulti-step tasks with tools: support desks, coding agents, function calls 129 of 160 not measured
Reasoning and knowledgeHard questions across science, maths and general knowledge 132 of 188 113 of 187
WritingWhat people prefer in blind comparisons, and judged writing quality 90 of 170 90 of 169
Sticking to the factsSummarising without inventing things, and factual answers 104 of 140 not measured

Every shared result

Each model's best result on every measure both have, grouped by benchmark, with the setting that produced it. Under each measure, every model that benchmark has measured is a grey tick, best to the right, so you can see where the two sit in the field. The headline measures of each benchmark come first; the rest are under "Show all". Green marks the better value; where a source reports ranges that overlap, neither is marked.

MeasureDeepSeek-R1DeepSeek-R1-0528
Arena (formerly LMArena) reported by Arena (formerly LMArena) · 3 results
Overall · TextArena rating, higher is better 141 · 110 of 387 measured 1,398 1,422
Business, management and finance · TextArena rating, higher is better 151 · 130 of 380 measured 1,388 1,407
Creative writing · TextArena rating, higher is better 125 · 109 of 385 measured 1,374 1,390
Artificial Analysis reported by Artificial Analysis · 3 results
Artificial Analysis Intelligence Indexindex score, higher is better 163 · 146 of 249 measured 11.4 13.1
GPQA Diamond% of questions, higher is better 155 · 104 of 232 measured 70.8% 81.3%
Humanity's Last Exam% of questions, higher is better 170 · 131 of 250 measured 8.5% 15.8%
UGI Leaderboard reported by UGI Leaderboard · 3 results
Writing score · UGIscore out of 100, higher is better 74 · 100 of 230 measured 51.8 42.3
Requested-length error · UGI% off the requested word count, lower is better 91 · 200 of 230 measured 18.0% 45.0%
Style adherence · UGIscore from 0 to 1, higher is better 203 · 183 of 230 measured 0.30 0.32
Show all 16 results (7 more)
MeasureDeepSeek-R1DeepSeek-R1-0528
Arena (formerly LMArena) reported by Arena (formerly LMArena) · 3 results
Expert prompts · TextArena rating, higher is better 151 · 136 of 338 measured 1,402 1,418
Instruction following · TextArena rating, higher is better 128 · 135 of 387 measured 1,397 1,391
Writing, literature and language · TextArena rating, higher is better 130 · 122 of 386 measured 1,382 1,389
Artificial Analysis reported by Artificial Analysis · 4 results
IFBench% of instructions, higher is better 141 · 137 of 195 measured 39.0% 39.6%
Long-context reasoning (AA-LCR)% of questions, higher is better 148 · 150 of 241 measured 57.7% 55.7%
Terminal-Bench Hard% of tasks, higher is better 147 · 107 of 190 measured 6.1% 15.9%
τ²-bench telecom% of tasks, higher is better 182 · 119 of 193 measured 11.4% 36.6%

Quick answers

Is DeepSeek-R1 better than DeepSeek-R1-0528?

DeepSeek-R1-0528 is the newer version; they score about the same at about the same price.

Which is better for reasoning and knowledge, DeepSeek-R1 or DeepSeek-R1-0528?

DeepSeek-R1-0528 ranks 113 of 188 models on sale for hard questions across science, maths and general knowledge, against 132 for DeepSeek-R1.

Run DeepSeek-R1 and DeepSeek-R1-0528 on your own prompt.

Benchmarks aren't your task. Test both side by side on your prompt, with every answer checked against your rules and the cost of each one.