DeepSeek R1

DeepSeekreasoningOpen Source

DeepSeek's reasoning model. Competitive with o3 on math and coding at a fraction of the cost.

Context Window

128K

tokens

Max Output

8K

tokens

Input Price

$0.55

per 1M tokens

Output Price

$2.19

per 1M tokens

Speed

35

tokens/sec

Released

Jan 2025

2025-01-20

Blended Cost

$1.37

per 1M tokens

Value Score

66.4

quality per $

Capabilities

ChatCode GenerationReasoning

Benchmarks

Quality Index
91
MMLU Pro
85.9
HumanEval (Coding)
91.6
MATH
94.3
Arena ELO
1350

Open Source — Licensed under MIT

About DeepSeek R1

DeepSeek R1 is a reasoning AI model by DeepSeek, released on January 20, 2025. It supports a context window of 128K tokens and can generate up to 8K output tokens.

At $0.55 per million input tokens and $2.19 per million output tokens, its blended cost of $1.37/1M tokens puts it in the mid-range pricing tier. Its value score of 66.4 reflects the balance of quality and cost.

DeepSeek R1 is available as an open-source model under the MIT license, meaning you can self-host it for predictable costs or use it through API providers like Swfte Connect.

Using DeepSeek R1 with Swfte

Access DeepSeek R1 through Swfte Connect, our unified LLM gateway. Connect gives you a single API for 50+ models, with automatic routing, cost optimization, and fallback handling. You can also try DeepSeek R1 in our AI Playground before integrating.