DeepSeek-R1: Benchmarks, Pricing & Context Window
DeepSeek-R1 is a language model from DeepSeek, released in January 2025.
DeepSeek-R1 is a reasoning-focused language model from DeepSeek that features advanced thinking capabilities. It serves as the foundation for DeepSeek's reasoning model family and pioneered their thinking mode approach for complex
DeepSeek-R1 pricing
Providers
DeepSeek-R1 starts at $0.550 per million input tokens and $2.19 per million output tokens via DeepSeek. See all 4 providers below with their per-token pricing, latency, throughput, and modality support.
| Provider | Input $/M | Cached input $/M | Output $/M | Context in / out | TTFT p95 s | Output p5 c/s | Modalities in / out |
|---|---|---|---|---|---|---|---|
 DeepSeek | $0.550 | — | $2.19 | 131.1K/131.1K | 0.30 | — | / |
 DeepInfra | $0.850 | — | $2.50 | 131.1K/131.1K | 0.30 | — | / |
 Together | $7.00 | — | $7.00 | 131.1K/131.1K | 0.60 | — | / |
 Fireworks | $8.00 | — | $8.00 | 131.1K/131.1K | 0.30 | — | / |
Cached input is the discounted price for prompt tokens served from a provider cache. TTFT is time to first token. Output is characters per second; p5 is the sustained floor exceeded by 95% of observed requests.
DeepSeek-R1 model size
DeepSeek-R1 has 671 billion parameters. See how it compares to other models in the same parameter range.
DeepSeek-R1 context window
Input and output token limits for DeepSeek-R1, plus how it ranks on long-context understanding.
Try now
huggleMake it with
DeepSeek-R1.
DeepSeek-R1 latency
DeepSeek-R1 time to first token, sustained output throughput, and failed-request rate from live model usage over the trailing 7 days.
DeepSeek-R1 examples
Recent arena outputs from DeepSeek-R1, picked from the highest-ranked matchups.
DeepSeek-R1 license
DeepSeek-R1 is released under the MIT license, which permits commercial use, has 671.0B parameters.
- License
- MIT
- Commercial use allowed
- Parameters
- 671.0B
MIT License - allows commercial use
DeepSeek-R1 resources
Official sources for DeepSeek-R1: provider documentation, official playground, official launch post, model weights.
FAQ
Common questions about DeepSeek-R1.