LLM Reference

DeepSeek R1

Released
2025-01-20
Last refreshed
2026-06-29
Status
Researched 90d ago
Open sourceCommercial use: permittedCodingRAGAgentsLong contextClassificationJSON / Tool use

DeepSeek R1 is worth evaluating for coding, rag, and agents when its provider route and context window match the workload.

Use it for

  • Teams evaluating coding, rag, and agents
  • Workloads that can use a 128k context window
  • Buyers comparing 4 tracked provider routes

Do not use it for

  • Vision or document-understanding workloads
Specifications
Released
2025-01-20
Context
128k
Parameters
671B, 37B Active
Architecture
Decoder Only
Knowledge cutoff
2023-12
Specialization
general
Openness
Open source
License
MITOSI-approvedCommercial use: permitted
Weights
Available
Code
Unknown
Training
Multi-stage
Created by

Advancing artificial general intelligence (AGI).

Hangzhou, Zhejiang, China
Founded 2023
Website
Pricing
Output / 1M
$0.300
Input / 1M
$0.100

Cheapest of 14 routes · Bitdeer AI

About

DeepSeek R1: Reasoning-optimized model with extended thinking capabilities. 128K context.

DeepSeek R1 is an open-source model. The structured metadata tracks a 128k-token context window, reasoning, structured outputs, and code execution. This page tracks provider routes through DeepSeek Platform, OpenRouter, Together AI, and 11 more, with the cheapest tracked route listed at $0.1 input and $0.3 output per 1M tokens. Headline tracked benchmarks include HumanEval 89.9, SWE-bench Verified 49.2, and Aider Polyglot 56.9.

Top use-case fit: coding, agents, and build tasks

Coding

Q/$ B

3 relevant benchmarks in the decision map.

RAG

Included by capability and metadata signals in the decision map.

Agents

Q/$ A

1 relevant benchmark in the decision map.

Provider price ladder

Compare all 14

Compare API pricing across 4 providers for input and output tokens, batch, and cached reads when available.

ProviderInput / 1MOutput / 1MRoute
Bitdeer AI$0.100$0.300
Serverless
SiliconFlow$0.250$0.800
Serverless
Fireworks AI$0.560$1.68
Serverless
DeepSeek Platform$0.550$2.19
Serverless

Available via routers & gateways(16)

LiteLLM

Gateway

Open-source Python SDK and proxy server that unifies 100+ LLM APIs behind a single OpenAI-compatible interface, with load balancing, cost tracking, and configurable failover.

Free OSSGCP Vertex AIMicrosoft FoundryDeepSeek Platform

OpenRouter

Hybrid

Unified hybrid gateway to 400+ models from 60+ providers via a single OpenAI-compatible API, with optional auto-routing that selects the best model per prompt.

PassthroughGCP Vertex AIDeepSeek PlatformTogether AI+1

Portkey

Gateway

Production AI gateway routing to 1,600+ LLMs with failover, load balancing, semantic caching, and guardrails; Apache 2.0 core is fully self-hostable with the complete feature set.

SubscriptionGCP Vertex AIMicrosoft Foundry

AIRouter

Router

Commercial LLM router that analyzes incoming requests and routes to the optimal model for cost/quality/latency via a drop-in OpenAI-compatible API, with a privacy-preserving embedding mode that avoids sending prompt content.

Passthrough + feeGCP Vertex AI

Amazon Bedrock Intelligent Prompt Routing

Router

AWS Bedrock's native intelligent prompt router that routes prompts between Anthropic Claude model tiers (Haiku/Sonnet) based on predicted task complexity, with no extra per-routing charge.

PassthroughAWS Bedrock

Azure AI Foundry Model Router

Router

Microsoft Azure AI Foundry's native model router that uses a trained ML model to route each prompt in real time to the optimal Azure-hosted model, with Balanced/Cost/Quality mode selection and automatic failover.

PassthroughMicrosoft Foundry

Capabilities

ReasoningStructured OutputsCode Execution

Benchmark peer barsfor Coding

Benchmark scores(5)

Scores are benchmark-specific and are direction-aware: the same numeric gap can mean very different outcomes across suites. Use the leaderboard context and this model's provider route to decide whether the winning margin is meaningful for your workload.
BenchmarkScoreVersionEvaluationSource
HumanEval89.92025-01Observed 2026-04-09Source
SWE-bench Verified49.22025-01Observed 2026-04-09Source
Aider Polyglot56.92026-04Observed 2026-04-14Source
Chatbot Arena1372.0Observed 2026-04-15Source
Google-Proof Q&A71.5diamondObserved 2026-04-15Source

Migration checks

No linked migration route is available for this model yet.

Compare DeepSeek R1 with other models

Show all 71 popular comparisonssorted by 7-day search impressions
DeepSeek R1 vs DeepSeek V3.11KDeepSeek R1 vs Qwen3-235B-A22B1KDeepSeek R1 vs Qwen3-Max1KDeepSeek R1 vs Gemini 2.5 Flash836DeepSeek R1 vs Qwen3.6-27B654DeepSeek R1 vs DeepSeek V4 Flash602DeepSeek R1 vs Qwen3.6-35B-A3B560DeepSeek R1 vs Qwen2.5-72B-Instruct522DeepSeek R1 vs Claude Sonnet 4.6460DeepSeek R1 vs GLM-5.1437DeepSeek R1 vs Xiaomi MiMo-V2.5-TTS-Series420DeepSeek R1 vs Xiaomi MiMo-V2.5348DeepSeek R1 vs Claude Opus 4.7303DeepSeek R1 vs Gemini 2.5 Flash Live API294DeepSeek R1 vs Kimi K2 Thinking289DeepSeek R1 vs Hunyuan Hy3 Preview281DeepSeek R1 vs GLM-5269DeepSeek R1 vs Claude Haiku 4.5243DeepSeek R1 vs GPT-5.5193DeepSeek R1 vs Qwen2.5-72B191DeepSeek R1 vs Claude Opus 4.5162DeepSeek R1 vs Kimi K2.5130DeepSeek R1 vs Qwen3.5-27B119DeepSeek R1 vs Ling-2.6-1T111DeepSeek R1 vs Tencent Hunyuan Turbo S110DeepSeek R1 vs Claude Sonnet 4.5109DeepSeek R1 vs Ling-2.6-Flash95DeepSeek R1 vs Llama 3 8B Instruct94DeepSeek R1 vs GPT-5.490DeepSeek R1 vs Grok 4.384DeepSeek R1 vs Trinity-Large-Thinking82DeepSeek R1 vs GPT-4o-mini Search Preview57DeepSeek R1 vs Llama 3.2 1B54DeepSeek R1 vs Step 3.5 Flash50DeepSeek R1 vs Qwen3-9B45DeepSeek R1 vs DeepSeek R1 052840DeepSeek R1 vs Llama 3.1 70B Instruct40DeepSeek R1 vs Mistral Large 239DeepSeek R1 vs Qwen3.6 Max Preview38DeepSeek R1 vs Grok 3 Mini38DeepSeek R1 vs Phi-3 Mini 4k37DeepSeek R1 vs Llama 3 70B Instruct35DeepSeek R1 vs GPT-5.230DeepSeek R1 vs Qwen2.5-7B-Instruct28DeepSeek R1 vs GLM-5 9B26DeepSeek R1 vs Llama 2 13B Chat21DeepSeek R1 vs Mistral Large 3 675B Instruct20DeepSeek R1 vs Gemma 7B Instruct19DeepSeek R1 vs o3 Deep Research17DeepSeek R1 vs Trinity-Large-Preview14DeepSeek R1 vs Qwen3.5-122B-A10B14DeepSeek R1 vs GLM-5V-Turbo12DeepSeek R1 vs Mixtral 8x22B v0.112DeepSeek R1 vs Llama 3.2 1B Instruct11DeepSeek R1 vs Qwen3.5-397B-A17B11DeepSeek R1 vs Together AI - Llama 3 8B Lite10DeepSeek R1 vs Together AI Qwen2-72B-Instruct7DeepSeek R1 vs DeepSeek R1 Distill Llama 70B7DeepSeek R1 vs GLM-5 Turbo7DeepSeek R1 vs Mixtral 8x22B Instruct v0.37DeepSeek R1 vs Qwen3.5-35B-A3B6DeepSeek R1 vs GPT-5.4-Cyber6DeepSeek R1 vs Gemma 2 2B5DeepSeek R1 vs Gemma 2B Instruct5DeepSeek R1 vs Qwen2-7B-Instruct4DeepSeek R1 vs Mistral Nemotron4DeepSeek R1 vs Phi-4 Mini Flash Reasoning4DeepSeek R1 vs Gemma 2 9B SahabatAI Instruct4DeepSeek R1 vs GPT-4o Search Preview2DeepSeek R1 vs Together AI Qwen2-7B-Instruct1DeepSeek R1 vs Mixtral 8x7B0

Frequently asked questions

What is the context window of DeepSeek R1?

DeepSeek R1 has a context window of 128k tokens.

How much does DeepSeek R1 cost?

DeepSeek R1 pricing ranges from $0.1/1M to $3.75/1M input tokens depending on the provider.

When was DeepSeek R1 released?

DeepSeek R1 was released on 2025-01-20.

Which providers offer DeepSeek R1?

DeepSeek R1 is available from 14 providers: DeepSeek Platform, OpenRouter, Together AI, Fireworks AI, NVIDIA NIM, Microsoft Foundry, Databricks Foundation Model Serving, DeepInfra, AWS Bedrock, Bitdeer AI, SiliconFlow, Replicate API, GCP Vertex AI, Vercel AI Gateway.

What benchmarks has DeepSeek R1 been tested on?

DeepSeek R1 has been evaluated on 5 benchmarks, including HumanEval, SWE-bench Verified, Aider Polyglot, Chatbot Arena, Google-Proof Q&A.