deepseek-reasoner
- Input / 1M
- $0.28
- Output / 1M
- $0.42
- Cached input
- $0.028
- Context
- 131K
A 1,000-token prompt costs $0.00028 to send. Token counts for DeepSeek models are estimated — approximated, because this family has no public tokenizer.
- Reasoning
- Max output 66K
Priced closest to this
| Model | Provider | Input | Output | |
|---|---|---|---|---|
| deepseek-chat | DeepSeek | $0.28 | $0.42 | |
| deepseek-v3.2 | DeepSeek | $0.28 | $0.40 | |
| deepseek-v3 | DeepSeek | $0.27 | $1.10 | |
| command-light | Cohere | $0.30 | $0.60 | |
| gemini-2.5-flash-native-audio-latest | $0.30 | $2.50 | ||
| gemini-2.5-flash-native-audio-preview-09-2025 | $0.30 | $2.50 |