# deepseek-v4-flash — API, Pricing & Context Window

`deepseek-v4-flash` is the fast, ultra-affordable member of the DeepSeek V4 family. It keeps the line's standout **1M-token context window** and **384K-token max output** while pricing input and output tokens at a fraction of frontier models.

Vivgrid serves `deepseek-v4-flash` through its unified, OpenAI-compatible API, making it a compelling default for high-volume, cost-sensitive workloads.

## Specifications

|     |     |
| --- | --- |
| **Provider** | DeepSeek |
| **Model ID** | `deepseek-v4-flash` |
| **Best for** | General-purpose |
| **Context window** | 1,000,000 tokens |
| **Max output** | 384,000 tokens |
| **Modalities** | Text |
| **Tool / function calling** | Yes |
| **Knowledge cutoff** | 2025-05 |
| **Acceleration** | 🌐 Global (Centralized) |

## Pricing

Pricing in USD per 1M tokens, matching the provider's rates.

| Input | Cached input | Output |
| --- | --- | --- |
| $0.15 | $0.03 | $0.28 |

## Quick start

Call `deepseek-v4-flash` through Vivgrid's unified, OpenAI-compatible endpoint. Get an API key from the [Vivgrid Console](https://console.vivgrid.com/).

```
curl https://api.vivgrid.com/v1/chat/completions \
  -H "Authorization: Bearer $VIVGRID_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{\n    "model": "deepseek-v4-flash",\n    "messages": [\
      { "role": "user", "content": "Say hello in English, Chinese and Spanish." }\
    ],\n    "stream": true\n  }'\n```

## Ideal use cases

- Very high-volume, cost-sensitive agent traffic
- Long-context summarization and extraction
- Large-output generation at minimal cost
- First-pass steps in multi-model pipelines

## Related models

- [deepseek-v4-pro](/content/docs/models/deepseek-v4-pro/index.html) — the flagship V4 model
- [deepseek-v3.2](/content/docs/models/deepseek-v3.2) — prior-generation model
- [gpt-5.4-nano](/content/docs/models/gpt-5.4-nano/index.html) — comparable ultra-low-cost option