Skip to content

DeepSeek · Chat

DeepSeek R1

DeepSeek 🇨🇳ChatDeepSeek familyOne of its sources says the weights aren't publicly available.The maker lists this model as deprecated (13 Aug 2026).
From Neosantara's listing:
DeepSeek-R1 is a state-of-the-art reasoning model, providing high-quality intelligence with excellent cost-efficiency.
llminfo IQ
121.9
Maker's price
$0.96 blended / 1M
Max context
1M
Listings
96

Evidence and sources

🤗 deepseek-ai/DeepSeek-R1-0528llminfo IQ 121.9, ranked among chat models. From 412 benchmarks. Plenty of data behind it.How llminfo IQ is computedCoding IQ 92.9, 95% interval 79–107. From 16 benchmark families across 12 sources. Plenty of data behind it. Uses the model's best published results. It has its own scale, separate from llminfo IQ.How Coding IQ is computedPoor value (36.1). Costs more per task than its benchmark results predict. From 36 benchmarks with measured task costs. Plenty of data behind it.How Value for money worksYap score: tokens used per coding task vs. models with the same resultsChatty (60 ±11). Generates more tokens than models with similar coding results. Measured at default effort. A fair amount of data behind it.Median 30,147 output tokens per task at default effort.How the Yap score worksReleased 20 Jan 2025 Training-data cutoff, as its sources state it Maker HQ: CN (according to DeepSeek's own website)
All rankings: #1 in X discussions today #2 in Wikipedia pageviews this week #13 in YouTube mentions this week #33 in Bluesky discussions this week #87 on NanoGPT recently #205 on OpenRouter today #218 on OpenRouter this week #221 on OpenRouter this monthAll rankings: #1 in X discussions today #2 in Wikipedia pageviews this week #13 in YouTube mentions this week #33 in Bluesky discussions this week #87 on NanoGPT recently #205 on OpenRouter today #218 on OpenRouter this week #221 on OpenRouter this monthSources: LMArena data, SiliconFlow data, LiteLLM data, Models.dev data, AI/ML API data, OpenRouter data docs, Neosantara Inference data, llminfo researchPrimary data from LMArena from SiliconFlow: name from LiteLLM: deprecation date from Models.dev: weights availability from AI/ML API: primary mode from OpenRouter: knowledge cutoff from Neosantara Inference: description from llminfo research: release date
Tags from its sources:open-sourcereasoninglong-context

Advantages and trade-offs

Compared with 10 chat models, picked mainly by popularity and IQ

Trade-offs

Deprecated by its maker

The maker has already deprecated this model (13 Aug 2026).

No open weights

The weights aren't public, so you can't run it yourself.

We only make a claim when there's enough data on models of the same type. Updated 22 Nov, 06:07 UTC.Prices come from the model's maker where they sell it, otherwise their official partner, otherwise the market median. This comparison uses 9 maker's prices; 1 market median price.

Input & output

as stated by the model's providers
Text
Text

Capabilities

Supported operations and provider-stated features

ReasoningFunction callingStructured outputPrompt cachingBatch API

Listings

 DeepSeekdeepseek/deepseek-r1Source: LiteLLM data66K$0.55cache read $0.14$2.19
 Flowdotdeepseek/deepseek-r1-0528:free · free, 0528Source: FlowDot data164K$0.00$0.00
 OpenRouterdeepseek/deepseek-r1:free · freeSource: Helicone Costs dataNot published by the provider$0.00$0.00
 Lambda AIlambda_ai/deepseek-r1-0528 · 0528Source: LiteLLM data131K$0.20$0.60
 Submodeldeepseek-ai/DeepSeek-R1-0528 · 0528Source: Models.dev data75K$0.20$0.80
 Eden AIInference by DeepInfradeepinfra/deepseek-ai/DeepSeek-R1-0528 · 0528Source: Models.dev data164K$0.50cache read $0.35$2.15
 Fastrouterdeepseek-ai/DeepSeek-R1Source: FastRouter data164K$0.50$2.18
 Hugging FaceInference by DeepInfradeepseek-ai/DeepSeek-R1-0528 · 0528Source: Hugging Face Inference Providers data164K$0.50$2.15
 Meganovadeepseek-ai/DeepSeek-R1-0528 · 0528Sources: MegaNova data, Models.dev data164K$0.50cache read $0.40$2.15
 NanoGPTdeepseek-ai/DeepSeek-R1-0528 · 0528Sources: Models.dev data, NanoGPT164K$0.50cache read $0.35$2.15

8 listings without a price are hidden.

Benchmarks

Named effort levels run from lowest to highest. Select a result to see its source details.

Math

AIME 2024Source: LLM Stats data · Reported by the model's developer. The source relaying it did not verify it. claim
AIME 2025Source: LLM Stats data · Reported by the model's developer. The source relaying it did not verify it. claim
HMMT 2025Source: LLM Stats data · Reported by the model's developer. The source relaying it did not verify it. claim
MathArenaSource: MathArena data
mu-MATH7 related benchmarks · 35 results
F1
Source: U-MATH and mu-MATH
F1 · Qwen/Qwen2.5-72B-Instruct
accuracy
Source: U-MATH and mu-MATH
accuracy · Qwen/Qwen2.5-72B-Instruct
MCC
Source: U-MATH and mu-MATH
MCC · Qwen/Qwen2.5-72B-Instruct
NPV
Source: U-MATH and mu-MATH
NPV · Qwen/Qwen2.5-72B-Instruct
PPV
Source: U-MATH and mu-MATH
PPV · Qwen/Qwen2.5-72B-Instruct
TNR
Source: U-MATH and mu-MATH
TNR · Qwen/Qwen2.5-72B-Instruct
Show 5 more resultsHide 5 results
TPR
Source: U-MATH and mu-MATH
TPR · Qwen/Qwen2.5-72B-Instruct
U-MATH7 related benchmarks · 21 results
algebra accuracy
Source: U-MATH and mu-MATH
Task set · algebra · gpt-4o-2024-08-06
differential_calc accuracy
Source: U-MATH and mu-MATH
Task set · differential_calc · gpt-4o-2024-08-06
integral_calc accuracy
Source: U-MATH and mu-MATH
Task set · integral_calc · gpt-4o-2024-08-06
multivariable_calculus accuracy
Source: U-MATH and mu-MATH
Task set · multivariable_calculus · gpt-4o-2024-08-06
overall accuracy
Source: U-MATH and mu-MATH
Task set · u_math · gpt-4o-2024-08-06
precalculus_review accuracy
Source: U-MATH and mu-MATH
Task set · precalculus_review · gpt-4o-2024-08-06
Show 3 more resultsHide 3 results
sequences_series accuracy
Source: U-MATH and mu-MATH
Task set · sequences_series · gpt-4o-2024-08-06
ValsSource: Vals AI

Safety

HELM66 related benchmarks · 73 results
Show 66 more resultsHide 66 results
Safety: BBQ - BBQ (ambiguous)
Source: Stanford HELM specialized suites data docs
Setting not stated
Setting not stated
Safety: BBQ - BBQ (unambiguous)
Source: Stanford HELM specialized suites data docs
Setting not stated
Setting not stated
Safety: XSTest - LM Evaluated Safety score
Source: Stanford HELM specialized suites data docs
Setting not stated
Setting not stated
PhareSource: Phare

Coding

LiveCodeBench11 related benchmarks · 12 results
LiveCodeBench
Source: LiveCodeBench Leaderboard data
Source: LLM Stats data · Reported by the model's developer. The source relaying it did not verify it. claim
Setting not stated
AtCoder
Source: LiveCodeBench Leaderboard data
Codeforces
Source: LiveCodeBench Leaderboard data
platform:codeforces
Easy
Source: LiveCodeBench Leaderboard data
Hard
Source: LiveCodeBench Leaderboard data
LeetCode
Source: LiveCodeBench Leaderboard data
Show 5 more resultsHide 5 results
Medium
Source: LiveCodeBench Leaderboard data
Pro
Source: LiveCodeBench Pro data · Oct 2025
Multi-SWE-bench10 related benchmarks · 28 results
C Verified
Source: Multi-SWE-bench Experiments data docs · Mar 2025
MagentLess, 20250329_MagentLess_DeepSeek-R1
MopenHands, 20250329_MopenHands_DeepSeek-R1
MSWE-agent, 20250329_MSWE-agent_DeepSeek-R1
C++ Verified
Source: Multi-SWE-bench Experiments data docs · Mar 2025
MagentLess, 20250329_MagentLess_DeepSeek-R1
MopenHands, 20250329_MopenHands_DeepSeek-R1
MSWE-agent, 20250329_MSWE-agent_DeepSeek-R1
Flash All Verified
Source: Multi-SWE-bench Experiments data docs · Jul 2025
MopenHands, 20250712_MopenHands_DeepSeek-R1-0528
Go Verified
Source: Multi-SWE-bench Experiments data docs · Mar 2025
MagentLess, 20250329_MagentLess_DeepSeek-R1
MopenHands, 20250329_MopenHands_DeepSeek-R1
MSWE-agent, 20250329_MSWE-agent_DeepSeek-R1
Java Verified
Source: Multi-SWE-bench Experiments data docs · Mar 2025
MagentLess, 20250329_MagentLess_DeepSeek-R1
MopenHands, 20250329_MopenHands_DeepSeek-R1
MSWE-agent, 20250329_MSWE-agent_DeepSeek-R1
JavaScript Verified
Source: Multi-SWE-bench Experiments data docs · Mar 2025
MagentLess, 20250329_MagentLess_DeepSeek-R1
MopenHands, 20250329_MopenHands_DeepSeek-R1
MSWE-agent, 20250329_MSWE-agent_DeepSeek-R1
Show 12 more resultsHide 12 results
Lite All Verified
Source: Multi-SWE-bench Experiments data docs · Mar 2025
MagentLess, 20250329_MagentLess_DeepSeek-R1
MopenHands, 20250329_MopenHands_DeepSeek-R1
MSWE-agent, 20250329_MSWE-agent_DeepSeek-R1
Python Verified
Source: Multi-SWE-bench Experiments data docs · Mar 2025
MagentLess, 20250329_Agentless_DeepSeek-R1
MopenHands, 20250329_OpenHands_DeepSeek-R1
MSWE-agent, 20250329_SWE-agent_DeepSeek-R1
Rust Verified
Source: Multi-SWE-bench Experiments data docs · Mar 2025
MagentLess, 20250329_MagentLess_DeepSeek-R1
MopenHands, 20250329_MopenHands_DeepSeek-R1
MSWE-agent, 20250329_MSWE-agent_DeepSeek-R1
TypeScript Verified
Source: Multi-SWE-bench Experiments data docs · Mar 2025
MagentLess, 20250329_MagentLess_DeepSeek-R1
MopenHands, 20250329_MopenHands_DeepSeek-R1
MSWE-agent, 20250329_MSWE-agent_DeepSeek-R1
SWE-bench Verified9 related benchmarks · 9 results
SWE-bench Verified
Source: LLM Stats data · Reported by the model's developer. The source relaying it did not verify it. claim
SWE-bench Multilingual
Source: LLM Stats data · Reported by the model's developer. The source relaying it did not verify it. claim
Show 3 more resultsHide 3 results

Human preference

Arena ELO28 related benchmarks · 56 results
Arena ELO
Source: LMArena data · Nov 2026
Setting not stated
Arena-Hard-v2.0-Preview
Source: Arena-Hard-Auto data
Creative Writing, Ensemble GPT-4.1 and Gemini 2.5 as Judges (Best Configuration for Creative Writing)
Hard Prompt, Style Control, and Gemini-2.5 as Judge (Official Configuration)
Hard Prompt, Style Control, and GPT-4.1 as Judge (If prefer OpenAI API)
Chinese
Source: LMArena data · Nov 2026
Setting not stated
Creative Writing
Source: LMArena data · Nov 2026
Setting not stated
English
Source: LMArena data · Nov 2026
Setting not stated
Exclude Ties
Source: LMArena data · Nov 2026
Setting not stated
Show 43 more resultsHide 43 results
Expert
Source: LMArena data · Nov 2026
Setting not stated
French
Source: LMArena data · Nov 2026
Setting not stated
German
Source: LMArena data · Nov 2026
Setting not stated
Hard Prompts
Source: LMArena data · Nov 2026
Setting not stated
Hard Prompts English
Source: LMArena data · Nov 2026
Setting not stated
Industry Business and Management and Financial Operations
Source: LMArena data · Nov 2026
Setting not stated
Industry Entertainment and Sports and Media
Source: LMArena data · Nov 2026
Setting not stated
Industry Life and Physical and Social Science
Source: LMArena data · Nov 2026
Setting not stated
Industry Mathematical
Source: LMArena data · Nov 2026
Setting not stated
Industry Medicine and Healthcare
Source: LMArena data · Nov 2026
Setting not stated
Industry Software and It Services
Source: LMArena data · Nov 2026
Setting not stated
Industry Writing and Literature and Language
Source: LMArena data · Nov 2026
Setting not stated
Japanese
Source: LMArena data · Nov 2026
Setting not stated
Korean
Source: LMArena data · Nov 2026
Setting not stated
Longer Query
Source: LMArena data · Nov 2026
Setting not stated
Multi Turn
Source: LMArena data · Nov 2026
Setting not stated
Non English
Source: LMArena data · Nov 2026
Setting not stated
Polish
Source: LMArena data · Nov 2026
Russian
Source: LMArena data · Nov 2026
Setting not stated
Spanish
Source: LMArena data · Nov 2026
Setting not stated
Style-Controlled
Source: LMArena data · Nov 2026
Setting not stated

Agentic

hal-gaiaSource: HAL (Princeton)
Source: OpenRouter benchmark exports · Oct 2026
deepseek-r1-0528 · 90-day results
Across 2 providers lowest 47.6% median 53% highest 58.5%OpenRouter's automatic routing 50.3%
Show results from 2 providers and OpenRouter's automatic routingHide results from 2 providers and OpenRouter's automatic routing
  • Novita
  • OpenRouter's automatic routing
  • SiliconFlow

Knowledge

MMLU-Pro32 related benchmarks · 32 results
MMLU-Pro
Source: Hugging Face Official Benchmarks data · Reported by the model's developer. The source relaying it did not verify it. claim
MMLU-Redux
Source: LLM Stats data · Reported by the model's developer. The source relaying it did not verify it. claim
Show 26 more resultsHide 26 results

Science

GPQASource: LLM Stats data · Reported by the model's developer. The source relaying it did not verify it. claim
GPQA Diamond6 results
Source: Epoch AI Benchmarking Hub data
Setting not stated
Source: OpenRouter benchmark exports · Oct 2026
deepseek-r1-0528 · 90-day results
Across 4 providers lowest 65.2% median 78.9% highest 79.8%OpenRouter's automatic routing 76.4%
Show results from 4 providers and OpenRouter's automatic routingHide results from 4 providers and OpenRouter's automatic routing
  • SiliconFlow
  • StreamLake
  • DeepInfra
  • OpenRouter's automatic routing
  • Novita

Long context

Reasoning

General

Factuality

Instruction following

Multilingual

Speech

Vision

Other

MedHELM33 related benchmarks · 33 results
Nejumi 428 related benchmarks · 28 results
ALT_バイアス
Source: Nejumi Leaderboard 4 data docs
deepseek-ai/DeepSeek-R1-0528: reasoning-enabled
ALT_倫理・道徳
Source: Nejumi Leaderboard 4 data docs
deepseek-ai/DeepSeek-R1-0528: reasoning-enabled
ALT_制御性
Source: Nejumi Leaderboard 4 data docs
deepseek-ai/DeepSeek-R1-0528: reasoning-enabled
ALT_堅牢性
Source: Nejumi Leaderboard 4 data docs
deepseek-ai/DeepSeek-R1-0528: reasoning-enabled
ALT_毒性
Source: Nejumi Leaderboard 4 data docs
deepseek-ai/DeepSeek-R1-0528: reasoning-enabled
ALT_真実性
Source: Nejumi Leaderboard 4 data docs
deepseek-ai/DeepSeek-R1-0528: reasoning-enabled
Show 22 more resultsHide 22 results
AVG_jaster_0shot
Source: Nejumi Leaderboard 4 data docs
deepseek-ai/DeepSeek-R1-0528: reasoning-enabled
AVG_jaster_2shots
Source: Nejumi Leaderboard 4 data docs
deepseek-ai/DeepSeek-R1-0528: reasoning-enabled
GLP_アプリケーション開発
Source: Nejumi Leaderboard 4 data docs
deepseek-ai/DeepSeek-R1-0528: reasoning-enabled
GLP_コーディング
Source: Nejumi Leaderboard 4 data docs
deepseek-ai/DeepSeek-R1-0528: reasoning-enabled
GLP_一般的知識
Source: Nejumi Leaderboard 4 data docs
deepseek-ai/DeepSeek-R1-0528: reasoning-enabled
GLP_基礎的言語性能
Source: Nejumi Leaderboard 4 data docs
deepseek-ai/DeepSeek-R1-0528: reasoning-enabled
GLP_専門的知識
Source: Nejumi Leaderboard 4 data docs
deepseek-ai/DeepSeek-R1-0528: reasoning-enabled
GLP_応用的言語性能
Source: Nejumi Leaderboard 4 data docs
deepseek-ai/DeepSeek-R1-0528: reasoning-enabled
GLP_情報検索
Source: Nejumi Leaderboard 4 data docs
deepseek-ai/DeepSeek-R1-0528: reasoning-enabled
GLP_意味解析
Source: Nejumi Leaderboard 4 data docs
deepseek-ai/DeepSeek-R1-0528: reasoning-enabled
GLP_抽象的推論
Source: Nejumi Leaderboard 4 data docs
deepseek-ai/DeepSeek-R1-0528: reasoning-enabled
GLP_推論能力
Source: Nejumi Leaderboard 4 data docs
deepseek-ai/DeepSeek-R1-0528: reasoning-enabled
GLP_数学的推論
Source: Nejumi Leaderboard 4 data docs
deepseek-ai/DeepSeek-R1-0528: reasoning-enabled
GLP_構文解析
Source: Nejumi Leaderboard 4 data docs
deepseek-ai/DeepSeek-R1-0528: reasoning-enabled
GLP_知識・質問応答
Source: Nejumi Leaderboard 4 data docs
deepseek-ai/DeepSeek-R1-0528: reasoning-enabled
GLP_翻訳
Source: Nejumi Leaderboard 4 data docs
deepseek-ai/DeepSeek-R1-0528: reasoning-enabled
GLP_表現
Source: Nejumi Leaderboard 4 data docs
deepseek-ai/DeepSeek-R1-0528: reasoning-enabled
GLP_論理的推論
Source: Nejumi Leaderboard 4 data docs
deepseek-ai/DeepSeek-R1-0528: reasoning-enabled
GLP_関数呼び出し
Source: Nejumi Leaderboard 4 data docs
deepseek-ai/DeepSeek-R1-0528: reasoning-enabled
TOTAL_SCORE
Source: Nejumi Leaderboard 4 data docs
deepseek-ai/DeepSeek-R1-0528: reasoning-enabled
アラインメント(ALT)_AVG
Source: Nejumi Leaderboard 4 data docs
deepseek-ai/DeepSeek-R1-0528: reasoning-enabled
汎用的言語性能(GLP)_AVG
Source: Nejumi Leaderboard 4 data docs
deepseek-ai/DeepSeek-R1-0528: reasoning-enabled
ValsSource: Vals AI

Data credits: Aider LLM Leaderboards (Aider-AI), Apache-2.0 (licence) · Data by LLM Stats · Epoch AI Benchmarking Hub, CC BY 4.0 (licence) · LMArena, CC BY 4.0 (licence) · Arena-Hard-Auto, LMArena, Apache-2.0 (licence) · ArtifactsBench (Tencent Hunyuan), CC BY 4.0 (licence) · BullshitBench by Peter Gostev, MIT (licence) · EQ-Bench (Samuel J. Paech) — cite paech2023eqbench · ChemBench leaderboard, Jablonka group / lamalab-org. Public verified numeric result summaries; results licence not separately stated. · CSEBenchmark (Wang et al., IEEE S&P 2025). Copyright (c) 2024 NASP. MIT License. (licence) · FinSearchComp, CC BY 4.0 (licence) · Clinical Benchmarks (clinicalbenchmarks.ai), CC BY 4.0; fractions converted to percentages; source documents retain their own terms (licence) · OpenRouter public benchmark leaderboard and search-lane exports, CC BY 4.0. Fractional source scores converted to percentages; immutable dated snapshots. (licence) · HAL (Princeton) — cite the HAL paper · Stanford HELM (Stanford CRFM); scenario scores from Capabilities, Safety, Long Context, MedHELM, AIR-Bench, MMLU and Finance · JMLE2026-Bench and IgakuQA119 by Naoto Iwase; results licence not stated (code MIT, question data CC BY 4.0). · JudgeBench, Sijun Tan et al., ScalerLab · Hugging Face official benchmark leaderboards; licence varies by benchmark · LiveMCPBench leaderboard, ICIP-CAS (Mo et al., 2025). Sheet linked from the Apache-2.0 project README; sheet licence not separately stated. · LLM Chess by Maxim Saplin, Apache-2.0 (Copyright [2024] [Maxim Saplin]) (licence) · MathArena (ETH SRI / INSAIT), MIT — cite dekoninck2026matharena (licence) · MLE-bench (OpenAI), MIT (licence) · MMLU-ProX: Xuan et al., arXiv:2503.10497. Public project-page results; no results licence stated. · Toloka, U-MATH and mu-MATH public leaderboard. MIT Space; generated solution outputs retain underlying model terms and are not ingested. (licence) · Multi-SWE-bench experiments (multi-swe-bench/experiments), Apache-2.0; resolved-rate fractions converted to percentages (licence) · Nejumi Leaderboard 4, Weights & Biases Japan; public publisher-selected numeric evaluation results · Orak leaderboard, KRAFTON AI (Park et al., NeurIPS 2025). Public numeric scores; leaderboard licence not stated (code MIT not applied; CC BY-NC trajectories not used). · pfgen-bench, Preferred Networks. Apache-2.0 except LLM-generated outputs; aggregate scores only. (licence) · Phare (Giskard), CC BY 4.0 (licence) · Trishul Lab, PutnamBench public submitted results. Model identities and scaffold budgets preserved; result licence unstated. · Vectara Hallucination Leaderboard, Apache-2.0 (licence)

Also known as: azure_ai/deepseek-r1, crusoe/deepseek-ai/deepseek-r1-0528, deepinfra/deepseek-ai/deepseek-r1, deepinfra/deepseek-ai/deepseek-r1-0528, deepseek-ai/deepseek-r1, deepseek-ai/deepseek-r1-0528, deepseek-ai/deepseek-r1-0528-tput, deepseek-r1-0528, deepseek-r1-250528, deepseek-r1-aliyun, deepseek-r1-baidu, deepseek-r1-huoshan, deepseek-r1-huoshan-250528, deepseek.r1-v1:0, deepseek/deepseek-r1, deepseek/deepseek-r1-0528, deepseek/deepseek-r1-0528:free, deepseek/deepseek-r1:free, fireworks_ai/accounts/fireworks/models/deepseek-r1, fireworks_ai/accounts/fireworks/models/deepseek-r1-0528, gmi/deepseek-ai/deepseek-r1, gmi/deepseek-ai/deepseek-r1-0528, hyperbolic/deepseek-ai/deepseek-r1, hyperbolic/deepseek-ai/deepseek-r1-0528, lambda_ai/deepseek-r1-0528, nebius/deepseek-ai/deepseek-r1, nebius/deepseek-ai/deepseek-r1-0528, novita/deepseek/deepseek-r1, novita/deepseek/deepseek-r1-0528, pro/deepseek-ai/deepseek-r1, public/deepseek-r1, replicate/deepseek-ai/deepseek-r1, sambanova/deepseek-r1, siliconflow/deepseek-r1-0528, snowflake/deepseek-r1, sophnet/deepseek-r1, sophnet/deepseek-r1-0528, streamlake/deepseek-r1, tensorx/deepseek/deepseek-r1-0528, together_ai/deepseek-ai/deepseek-r1-0528, us.deepseek.r1-v1:0, vercel_ai_gateway/deepseek/deepseek-r1, vertex_ai/deepseek-ai/deepseek-r1-0528-maas

Like this one. Tap or hover any dotted value, label or icon to see where it comes from and how it is measured. open details when tapped or hovered. How this data works