# DeepSeek V4.1 Flash API

> DeepSeek's fast, low-cost model, with native image understanding. By DeepSeek, on Revana at $0.09 in / $0.36 out per 1M tokens.

- Model id: `deepseek-v4.1-flash`
- Kind: text (chat completions)
- Maker: DeepSeek
- Endpoint: `POST https://api.revana.si/v1/chat/completions` (OpenAI-compatible, streaming supported)
- Context window: 1M tokens
- Max output: 393K tokens
- Routes: 2 (starts on the cheapest, falls back automatically)
- Page: https://revana.si/llm/deepseek-v4.1-flash

## Price (USD per million tokens)

| Tokens | Price |
|---|---|
| Input | $0.09 |
| Cached input | $0.0018 |
| Cache write (5 min) | $0.09 |
| Cache write (1 hour) | $0.18 |
| Output | $0.36 |

Requests made weekdays 09:00-12:00 and 14:00-18:00 Beijing time (UTC+8) are billed at 2× these rates.
Reasoning tokens are billed as output. Each answer's `usage.cost_usd` is the amount taken from your balance; a request with no answer costs nothing.

## Request

```bash
curl https://api.revana.si/v1/chat/completions \
  -H "Authorization: Bearer $REVANA_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{"model":"deepseek-v4.1-flash","messages":[{"role":"user","content":"Write a haiku about routing."}]}'
```

Or any OpenAI SDK with `base_url="https://api.revana.si/v1"`. Docs: https://revana.si/chat