TokenCosts AI.iov1.0
Back to Models Directory
DeepSeekdeepseek-r1

DeepSeek-R1 (deepseek-reasoner) API Pricing

Official PricingCall via OpenRouter

Open-weights reasoning model with emergent reasoning capabilities. Reasoning tokens output rate is $2.19/1M. Automatic prefix caching yields ~74.5% input discount.

Base Input Price$0.55/ 1M tokens
Base Output Price$2.19/ 1M tokens
Prompt Cache Read$0.140-74.55% discount
Batch API RateN/A24h async offload

Inference Hosts & Aggregator Providers

Compare rates and latency across direct APIs, OpenRouter, and hyperscale cloud providers

Live verified rates
Provider NameInput / 1MOutput / 1MPrompt CachingAccess Endpoint
DeepSeek Direct$0.55$2.19SupportedConnect
OpenRouter$0.55$2.19SupportedConnect

Head-to-Head Comparisons for DeepSeek-R1 (deepseek-reasoner)

DeepSeek-R1 (deepseek-reasoner) vs OpenAI: GPT-6.1 Sol$0.55/M vs $2/M
DeepSeek-R1 (deepseek-reasoner) vs GPT-4o (Omni)$0.55/M vs $2.5/M
DeepSeek-R1 (deepseek-reasoner) vs Claude 3.5 Sonnet$0.55/M vs $3/M
DeepSeek-R1 (deepseek-reasoner) vs DeepSeek-V3 (deepseek-chat)$0.55/M vs $0.27/M