DeepSeek-R1: Pricing, Context, and Benchmarks

DeepSeek-R1 is DeepSeek's DeepSeek-R1 model released on 2025-01-20. It runs on a 128K-token context window, priced at $0.55/M input and $2.19/M output. Open-weight reasoning model competitive with o1/o3 on math and coding. Weights publicly downloadable.

Specifications

Benchmark scores

Scores reflect single measurements on the date shown. Reproducibility varies; consult the linked source for methodology. Use the comparison tool for side-by-side analysis.

What this model is good for

Based on benchmark scores and capability fit, DeepSeek-R1 ranks well for:

• Math & STEM Reasoning — rank #4. Best open-weight reasoning model. Much cheaper than commercial alternatives.
• Best Open-Weight Model — rank #4. Best open reasoning model for STEM tasks.

Cost calculator

For 1,000 tokens of input and 1,000 tokens of output per call, this model costs $0.0027 per call. For typical usage:

For more precise estimates including caching discounts, use the LLM API Cost Calculator.

Other DeepSeek models

• DeepSeek-V3

Common comparisons

• DeepSeek-R1 vs Llama 3.3 70B Side-by-side specs, pricing, and benchmarks
• DeepSeek-R1 vs Qwen3-235B Side-by-side specs, pricing, and benchmarks
• DeepSeek-R1 vs Gemini 2.5 Flash Side-by-side specs, pricing, and benchmarks
• DeepSeek-R1 vs GPT-5 Mini Side-by-side specs, pricing, and benchmarks

API setup

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

See the official documentation for the full request/response schema. For pre-flight token counting before paying, use the appropriate token counter.