← Models

DeepSeek R1

deepseek/deepseek-r1 · Reasoning-focused open-weight model

Try in Chat Compare Quickstart

Overview Routing Pricing Quickstart

Overview

DeepSeek R1 is optimized for chain-of-thought style reasoning and complex problem decomposition. Suitable for research assistants, code analysis, and multi-step workflows.

Languages: strong English and Chinese; extend via prompt design for Arabic and others.

Routing

We expose tiers rather than internal datacenter names:

  • Interactive tier — lower median latency for chat workloads.
  • Efficiency tier — cost-optimized pools for batch jobs.

Automatic failover applies when a pool returns 5xx or rate-limit errors.

Pricing

Indicative Basic rates: $0.20 / 1M input tokens, $0.40 / 1M output tokens. Pro volume discounts apply per Pricing.

Quickstart

curl https://api.tokenhub.example/v1/chat/completions \
  -H "Authorization: Bearer $TOKENHUB_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{"model":"deepseek/deepseek-r1","messages":[{"role":"user","content":"Hello"}]}'

Full docs →