DeepSeek · Text & reasoning · Via OpenRouter
DeepSeek V4 Flash Latest pricing
Text input: $0.03 / 1M tokens.
Source snapshot: · USD · Urbano DX
DeepSeek V4 Flash Latest: API pricing
| Charge | Published rate | Billing unit | Endpoint / tier |
|---|---|---|---|
| Text input | $0.03 | 1M tokens | Base |
| Text output | $0.07 | 1M tokens | Base |
| Cache read | $0.003 | 1M tokens | Base |
DeepSeek V4 Flash Latest: Cost calculator
Estimated text API cost
USD · Via OpenRouter
Input and cached input are separate quantities. The context check includes both plus output. Long-context rates apply automatically to the whole request above the listed threshold. Reasoning tokens belong in billable output. Cache writes, image/audio input, tools, taxes and top-up fees are outside this estimate. For scheduled pricing, the estimate uses the highest cost among the listed time windows.
Model details
- Exact model ID
- ~deepseek/deepseek-v4-flash-latest
- Developer
- DeepSeek
- Accepts
- Text
- Produces
- Text
- Context window (tokens)
- 1,310,720
- Maximum output (tokens)
- 393,216
Pricing sources and limits
All amounts are USD. Provider routing, service tier, region, request size and account terms can change the price. API usage is billed separately from consumer app subscriptions. Zero-price variants have provider limits and do not promise unlimited free use.
This directory covers the models and variants listed by OpenRouter, plus the fal H3 Max guide. OpenRouter rates are hosting prices, which can differ from buying directly from the developer. It is a dated snapshot, not a live quote or a complete list of every model in existence.
Compare other models
Discuss an AI integration
Turn the model budget into a working integration with cost controls and request tracing.
Discuss an AI integration →