DeepSeek: R1
DeepSeek R1 is here: Performance on par with [OpenAI o1](/openai/o1), but open-sourced and with fully open reasoning tokens. It's 671B parameters in size, with 37B active in an inference pass....
Everything else
| Max output | 16K |
|---|---|
| Input / 1M | $0.7 |
| Output / 1M | $2.50 |
| Cache read / 1M | — |
| Cache write / 1M | — |
| 1M in + 1M out | $3.20 |
| Takes | text |
| Returns | text |
| Tools | yes |
| Reasoning | always on |
| Open weights | yes |
| Released | 2025-01-20 |
| Knowledge cutoff | 2024-07-31 |
| Retires | — |
| Catalog id | deepseek/deepseek-r1 |
Benchmarks
| source | epoch |
|---|---|
| eci | 139.68 |
| confidence | Confident |
| accessibility | Open weights (unrestricted) |
| gpqa | 0.692 |
| gpqaStderr | 0.0307 |
| mathLevel5 | 0.931 |
| mathLevel5Stderr | 0.007 |
| aime | 0.533 |
| aimeStderr | 0.0752 |
Benchmark scores by Epoch AI, CC BY 4.0.
Using it
In 00, this model is picked per agent — and per tier, so the model that answers your customers need not be the one that writes your code. Get 00.