Model comparison
DeepSeek R1 vs DeepSeek V4 Pro
DeepSeek
DeepSeek R1
The open-weights reasoning model that reset the cost curve.
DeepSeek
DeepSeek V4 Pro
The price collapse - frontier quality at a fraction of the cost.
Specs
| Metric | DeepSeek R1 | DeepSeek V4 Pro |
|---|---|---|
| Context window | 64K tokens | 1.0M tokens↑ |
| Input $/1M tokens | $0.700 | $0.435↑ |
| Output $/1M tokens | $2.50 | $0.870↑ |
| Modalities | Text | Text |
| Open weights | Yes | Yes |
| Released | Jan 2025 | Dec 2024 |
Capability differences
| Capability | DeepSeek R1 | DeepSeek V4 Pro |
|---|---|---|
| Tool use | No | Yes |
| Extended thinking | Yes | No |
| Prompt caching | No | Yes |
Key differences
- -DeepSeek R1 offers 64K context; DeepSeek V4 Pro offers 1.0M.
- -Input pricing: DeepSeek R1 at $0.700/1M vs DeepSeek V4 Pro at $0.435/1M.