# deepseek-v3.1 — API, Pricing & Context Window

`deepseek-v3.1` is a reliable general-purpose model from DeepSeek with a **131K-token context window** and competitive pricing. It remains a solid option for established integrations and steady production workloads.

On Vivgrid, `deepseek-v3.1` is available through the unified, OpenAI-compatible endpoint, easing migration to newer DeepSeek releases when you're ready.

## Specifications

|     |     |
| --- | --- |
| **Provider** | DeepSeek |
| **Model ID** | `deepseek-v3.1` |
| **Best for** | General-purpose |
| **Context window** | 131,072 tokens |
| **Modalities** | Text |
| **Tool / function calling** | Yes |
| **Knowledge cutoff** | 2024-07 |
| **Acceleration** | 🌐 Global (Centralized) |

## Pricing

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

| Input | Cached input | Output |
| --- | --- | --- |
| $1.14 | — | $4.56 |

## Quick start

Call `deepseek-v3.1` 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 '{
    "model": "deepseek-v3.1",
    "messages": [\
      { "role": "user", "content": "Say hello in English, Chinese and Spanish." }\
    ],
    "stream": true
  }'
```

## Ideal use cases

- Established general-purpose agent integrations
- Cost-efficient summarization and extraction
- Steady, high-volume text workloads
- A baseline for comparing against V3.2 and V4 models

## Related models

- [deepseek-v3.2](/content/docs/models/deepseek-v3.2) — the next 3.x release
- [deepseek-v4-flash](/content/docs/models/deepseek-v4-flash/index.html) — modern low-cost model
- [deepseek-r1-0528](/content/docs/models/deepseek-r1-0528/index.html) — reasoning-focused model

### Additional models

- [deepseek-v3.2-exp](/content/docs/models/deepseek-v3.2-exp/index.html) — DeepSeek's experimental V3.2 model with a 164K context window.
- [deepseek-r1-0528](/content/docs/models/deepseek-r1-0528/index.html) — DeepSeek's reasoning-focused R1 model with a 164K context window.