DeepSeek
DeepSeek API
DeepSeek V4 — frontier reasoning at a fraction of Western frontier token prices.
2
models in the line
At a glance
- From
- $0.168 / 1M tokens
Models in this family
| Model | From | Modes | Open |
|---|---|---|---|
| DeepSeek V4 FlashFast, very low cost DeepSeek tier for high-volume classification, extraction and chat. | $0.168/ 1M tokens | — | OpenDocs |
| DeepSeek V4 ProDeepSeek's frontier reasoning model. Strong at maths, coding and long analytical chains at a fraction of Western frontier pricing. | $0.522/ 1M tokens | — | OpenDocs |
Pricing
| Model | Option | Price |
|---|---|---|
| DeepSeek V4 Flash | input | $0.168/ 1M tokens |
| DeepSeek V4 Flash | output | $0.336/ 1M tokens |
| DeepSeek V4 Pro | input | $0.522/ 1M tokens |
| DeepSeek V4 Pro | output | $1.04/ 1M tokens |
Which one to pick
DeepSeek V4 Flash
High-volume classification, extraction and chat where per-token cost dominates the decision.
DeepSeek V4 Pro
Maths, coding and long analytical chains — frontier reasoning without frontier pricing.
Strengths
- The cheapest frontier-class reasoning in the catalog on a per-token basis.
- Flash has the lowest output price of any text model in the catalog.
- Output is only two times input on both tiers, the narrowest ratio in the catalog and well below the GPT or Claude lines.
- OpenAI-protocol compatible, so it drops into existing code.
Limitations
- Only two tiers, with no mid-range option between them.
- Text-only — no image input in this family.
Frequently asked questions
How much does the DeepSeek API cost?
API Stock bills per use from one prepaid balance, with no subscription. Current prices: DeepSeek V4 Flash — $0.168 per 1M input tokens and $0.336 per 1M output tokens; DeepSeek V4 Pro — $0.522 per 1M input tokens and $1.04 per 1M output tokens.
Which DeepSeek model is the cheapest?
DeepSeek V4 Flash is the cheapest at $0.168 per 1M input tokens. The most expensive tier, DeepSeek V4 Pro, costs $0.522 per 1M input tokens.
Can I use the DeepSeek API with the OpenAI or Anthropic SDK?
Yes. DeepSeek models accept the OpenAI request format: set the SDK's base URL to https://api.api-stock.com/api/v1 and use your API Stock API key. Switching between models only changes the model id.
How do I get an API key for DeepSeek?
Create an account and API key on API Stock, then fund your balance. The same key works with DeepSeek and other catalog models.
The vendor's own description of this line: DeepSeek — API docs
Guides
- LLM API pricing: GPT, Claude, Gemini, DeepSeek comparedCompare LLM API pricing for GPT, Claude, Gemini, DeepSeek, Grok and Kimi per 1M tokens, with worked monthly cost examples and practical ways to cut the bill.9 min read
- Going to productionKey handling, webhooks over polling, idempotency, terminal states, retry policy by error code, balance monitoring, rate limits and media storage.10 min read