Open-weights frontier-style reasoning at low cost (MIT license; off-peak pricing; peak hours double rates; cache hit $0.022)
Stored values, not live provider measurements.
| Provider | DeepSeek |
| Category | Open-Weights |
| Input cost / 1M tokens | $0.660 |
| Output cost / 1M tokens | $1.98 |
| Cache read / 1M tokens | $0.022 |
| Context window | 1M tokens |
| Max output | 384k tokens |
| Batch API | Not listed |
| Vision | Text only |
| Vision price / 1k images | — |
| Speed rating | Moderate |
| Intelligence score | 23/100 |
| Throughput | 80 tokens/sec |
| Time to first token | 59.00s |
| Hardware | Open weights (Hugging Face) |
Enter your own input, output, caching, and batch assumptions in the calculator.
Open calculatorPricing review: 11/4/2026
Performance review: 11/4/2026
DeepSeek V4 Pro costs $0.660 per 1M input tokens and $1.980 per 1M output tokens in the stored catalog. Blended cost depends on your input/output mix.
DeepSeek V4 Pro has a context window of 1.0M tokens (1,000,000 tokens). This is the maximum combined input and output for a single request.
DeepSeek V4 Pro is currently offered by DeepSeek. Pricing is a stored catalog value — confirm current rates with the provider.