LV
LLMversus
DeepSeek

DeepSeek R1

Complete specs, pricing, and benchmark data for DeepSeek R1 by DeepSeek. Last verified 2026-04-03.

JSON ModeStreaming
Pricing

Input / 1M tokens

$0.550

Output / 1M tokens

$2.19

Cached Input / 1M

$0.140

Context & Output

Context Window

128K

Max Output

8,192

TTFT

1800ms

Speed

45 tok/s

Benchmarks

Arena ELO

1310

Coding ELO

1330

Reasoning ELO

1350

HumanEval

93

MMLU

89

MATH

91

GPQA

72

Price History (Input $/M tokens)

Strengths
  • +Excellent reasoning at a fraction of o3's cost
  • +Open-source and self-hostable
  • +Top-tier math performance
  • +Very competitive with proprietary reasoning models
Limitations
  • -High latency due to chain-of-thought
  • -No function calling support
  • -No vision/multimodal support

Best For

ReasoningMathCodingScience

Compare DeepSeek R1 with...

vs o4-minivs Gemini 2.5 Provs Claude Opus 4vs o3-minivs Grok 3vs GPT-4.1vs Llama 4 Maverickvs Claude Sonnet 4vs DeepSeek V3vs Gemini 2.0 Flashvs GPT-4ovs Qwen 2.5 Max
Official Pricing Page →