Models / DeepSeek-R1

DeepSeek

DeepSeek-R1

DeepSeek-R1 trails most of the field on what it has been measured on; it is not on sale through OpenRouter, so it is here for reference.

Superseded. DeepSeek-R1-0528 is DeepSeek's newer model in this line, at $0.92 blended: see DeepSeek-R1-0528 or compare the two.

Results 16 Nov 2026 · 23 results · 4 sources

Price per million tokens
Not available through OpenRouter
Speed
Not measured
Intelligence Index
11.4
Artificial Analysis
Spring Prompt overall
Not ranked yet
needs our benchmarks and two groups of results
Developer
DeepSeek, based in China
Weights not published
Released
20 Jan 2025

Where it stands among the models you could choose

Available models grouped by price, intelligence, speed and developer location. Retired models are excluded.

AmongIntelligence IndexSpring Prompt overallPrice, blendedSpeed
Models available today 130th of 186Claude Opus 5.5 – – –
Models from developers based in China 58th of 76MiMo-V2.6-Pro – – –

The name under each rank is the leader of that group; green is the top quarter of the group and red the bottom quarter, by rank. Price is shown as a percentile: the share of the group that costs less, so lower is cheaper. "Sets this group" marks the measure the group is defined by. Spring Prompt overall is our score out of 100 across every source (how it works). Price is the developer's list price (or the typical OpenRouter host where we haven't read one) for three input tokens to one output; intelligence and speed are from Artificial Analysis (data sourced from Artificial Analysis); "similar intelligence" means within 4 points on its Intelligence Index. Developer locations are where each company is based, not where a model is served. Groups under 3 models are not ranked.

How good is it, and for what?

Each use case ranks the available models its benchmarks measured. The bar shows its percentile in that field, best to the right. Open a row for the results behind it.

Product listingsTurning a sparse product feed and photos into listings that can go live Not measured
  • CatalogBench: has not measured this model.
Decks from an analysisTurning a finished analysis into a deck you could present as it is Not measured
  • DeckBench: has not measured this model.
User surveysPlanning a user survey and reading its results without being misled Not measured
  • SurveyBench: has not measured this model.
Marketing planningPlanning a year of ad spend without overspending Not measured
  • ROASBench: has not measured this model.
Agents and tool useMulti-step tasks with tools: support desks, coding agents, function calls 129th of 160 available
BenchmarkDeepSeek-R1Rank among availableBest available
Artificial AnalysisArtificial Analysis: Terminal-Bench 4.0 · reported 0.0% 67th of 108 Kimi K3.1 Pro 67.2%
Artificial AnalysisArtificial Analysis: Terminal-Bench 2.1 · reported 19.1% 82nd of 108 Claude Fable 5.1 91.4%
Artificial AnalysisArtificial Analysis: τ-bench banking · reported 6.4% 85th of 105 Grok 4.6 50.7%
  • tau2-bench: has not measured this model.
  • Berkeley Function Calling Leaderboard (BFCL) V4: has not measured this model.
  • OpenHands Index: has not measured this model.
  • Microsoft STATE-Bench: has not measured this model.
  • Vending-Bench 2: has not measured this model.
Professional workReal tasks from banking, consulting and law, business documents and freelance projects Not measured
  • APEX-Agents: has not measured this model.
  • GDP.pdf: has not measured this model.
  • Remote Labor Index: has not measured this model.
Reasoning and knowledgeHard questions across science, maths and general knowledge 132nd of 188 available
BenchmarkDeepSeek-R1Rank among availableBest available
Artificial AnalysisArtificial Analysis: Artificial Analysis Intelligence Index · reported 11.4 130th of 186 Claude Opus 5.5 57.6
Artificial AnalysisArtificial Analysis: Humanity's Last Exam · reported 8.5% 133rd of 187 Claude Opus 5.5 61.4%
Artificial AnalysisArtificial Analysis: GPQA Diamond · reported 70.8% 116th of 169 GPT-6 Astra 96.3%
WritingWhat people prefer in blind comparisons, and judged writing quality 90th of 170 available
BenchmarkDeepSeek-R1Rank among availableBest available
Arena (formerly LMArena)Arena Text: overall · reported 1,398 99th of 154 Gemini 4 Argon 1,525
UGI LeaderboardUGI: writing score · reported 51.8 56th of 125 Gemini 3.8 Flash 78.6
Sticking to the factsSummarising without inventing things, and factual answers 104th of 140 available
BenchmarkDeepSeek-R1Rank among availableBest available
Vectara Hallucination LeaderboardVectara: hallucination rate · reported 11.3% 52nd of 72 GPT-5.4 nano 3.1%
  • Arena (formerly LMArena): has not measured this model.
  • SimpleQA Verified (Epoch AI): has not measured this model.
Speed under pressureGood decisions against a real clock (fast chess) Not measured
  • BulletBench: has not measured this model.

Nearest alternatives

Against DeepSeek-R1, Intelligence Index 11.4. "Similar intelligence" means within 4 points or better.

Against the alternatives

The models you would most likely weigh it against: the leaders of the groups above.

Scroll sideways to see every alternative.

DeepSeek-R1Claude Opus 5.5
most intelligent available today
MiMo-V2.6-Pro
most intelligent from China
Claude Fable 5.2
leads overall
DeepSeek-V4.1-Flash
DeepSeek's best other model
Price per million tokens – $8.00$0.54$16.00$0.52
Tokens a second – 9643–369
Intelligence Index 11.4 57.646.3–39.5
Spring Prompt overall – 82–9051
Rank among available models, by use case
Product listings – 13th–5th15th
Decks from an analysis – 7th–4th–
User surveys – 4th–1st–
Marketing planning – 9th–1st–
Agents and tool use 129th of 160 3rd22nd–31st
Professional work – 3rd–––
Reasoning and knowledge 132nd of 188 1st13th–42nd
Writing 90th of 170 5th23rd–29th
Sticking to the facts 104th of 140 3rd10th–23rd
Speed under pressure – ––8th–

Shaded figures are better than DeepSeek-R1's. A dash means no result.

Where it has been measured

Compared with the models available today (it is not on sale itself), in the bottom quarter for agents and tool use.

Artificial Analysis lists it but has not published a speed for it yet

Every result

23 published results from 4 sources, each in the source's own units, with the configuration that produced it.

Arena (formerly LMArena) · reported by Arena (formerly LMArena) · 6 results
MeasureValueRankConfigurationDated
Arena Text: business, management and financeArena rating, higher is better 1,388
1,374–1,402
173 of 406 DeepSeek-R1 2 Oct 2026
Arena Text: creative writingArena rating, higher is better 1,374
1,364–1,384
148 of 411 DeepSeek-R1 2 Oct 2026
Arena Text: expert promptsArena rating, higher is better 1,402
1,382–1,422
172 of 363 DeepSeek-R1 2 Oct 2026
Arena Text: instruction followingArena rating, higher is better 1,397
1,390–1,405
148 of 413 DeepSeek-R1 2 Oct 2026
Arena Text: overallArena rating, higher is better 1,398
1,393–1,403
162 of 413 DeepSeek-R1 2 Oct 2026
Arena Text: writing, literature and languageArena rating, higher is better 1,382
1,374–1,391
152 of 412 DeepSeek-R1 2 Oct 2026

Not shown: Not accuracy or correctness: it ranks which answer voters preferred, with answer length and formatting controlled for.

Artificial Analysis · reported by Artificial Analysis · 12 results
MeasureValueRankConfigurationDated
Artificial Analysis: Artificial Analysis Coding Indexindex score, higher is better 24.6 143 of 187 DeepSeek-R1 20 Oct 2026
Artificial Analysis: Artificial Analysis Intelligence Indexindex score, higher is better 11.4 299 of 413 DeepSeek-R1 20 Oct 2026
Artificial Analysis: GPQA Diamond% of questions, higher is better 70.8% 246 of 360 DeepSeek-R1 20 Oct 2026
Artificial Analysis: Humanity's Last Exam% of questions, higher is better 8.5% 296 of 419 DeepSeek-R1 20 Oct 2026
Artificial Analysis: IFBench% of instructions, higher is better 39.0% 212 of 288 DeepSeek-R1 20 Oct 2026
Artificial Analysis: Long-context reasoning (AA-LCR)% of questions, higher is better 57.7% 257 of 407 DeepSeek-R1 20 Oct 2026
Artificial Analysis: SciCode% of problems, higher is better 38.3% 174 of 205 DeepSeek-R1 20 Oct 2026
Artificial Analysis: Terminal-Bench 2.1% of tasks, higher is better 19.1% 147 of 188 DeepSeek-R1 20 Oct 2026
Artificial Analysis: Terminal-Bench 4.0% of tasks, higher is better 0.0% 152 of 204 DeepSeek-R1 20 Oct 2026
Artificial Analysis: Terminal-Bench Hard% of tasks, higher is better 6.1% 224 of 282 DeepSeek-R1 20 Oct 2026
Artificial Analysis: τ-bench banking% of tasks, higher is better 6.4% 145 of 175 DeepSeek-R1 20 Oct 2026
Artificial Analysis: τ²-bench telecom% of tasks, higher is better 11.4% 275 of 286 DeepSeek-R1 20 Oct 2026

Not shown: Not your codebase or tools.

UGI Leaderboard · reported by UGI Leaderboard · 3 results
MeasureValueRankConfigurationDated
UGI: requested-length error% off the requested word count, lower is better 18.0% 158 of 379 DeepSeek-R1 21 Sep 2025
UGI: style adherencescore from 0 to 1, higher is better 0.30 341 of 379 DeepSeek-R1 21 Sep 2025
UGI: writing scorescore out of 100, higher is better 51.8 157 of 379 DeepSeek-R1 21 Sep 2025

Not shown: Not other format limits such as character counts or bullet counts.

Vectara Hallucination Leaderboard · reported by Vectara Hallucination Leaderboard · 2 results
MeasureValueRankConfigurationDated
Vectara: answer rate% of documents, higher is better 97.0% 88 of 108 DeepSeek-R1 22 Sep 2026
Vectara: hallucination rate% of summaries, lower is better 11.3% 74 of 108 DeepSeek-R1 22 Sep 2026

Not shown: Not a quality score: a low rate usually means content filters were triggered, and hallucination rates are measured on answered documents only.

Test your own prompt.

Benchmarks aren't your task. Run your prompt on the models you could use, with every answer checked and costed.