# DeepSeek V4 Pro

Canonical page: https://agentsky.dev/models/deepseek/deepseek-v4-pro
JSON record: https://gateway.agentsky.dev/v1/models/deepseek/deepseek-v4-pro

DeepSeek V4 Pro is a text-input model by DeepSeek, available through the AgentSky API as deepseek/deepseek-v4-pro at $0.924 per million input tokens and $2.772 per million output tokens, with a 1M context window.

The higher-capability V4 text model, with thinking mode on by default and a million-token window.

## Facts

- **Model id:** `deepseek/deepseek-v4-pro`
- **Maker:** DeepSeek
- **Input modalities:** Text
- **Context:** 1M
- **Max output tokens:** 384,000
- **Released:** Aug 13, 2026
- **Aliases:** `deepseek/deepseek-v4-pro`, `deepseek-v4-pro`

## Pricing

USD per million tokens on the default plan. Signed-in pages show your own plan's price; every response carries the cost of that call in `usage.cost`.

| Meter | Per 1M tokens |
|---|---|
| Input | $0.924 |
| Output | $2.772 |
| Cache read | $0.0308 |
| Cache write | $0.924 |

## Use it

Anthropic wire base URL: `https://gateway.agentsky.dev`. OpenAI wire base URL: `https://gateway.agentsky.dev/v1`. Authenticate with an AgentSky API key (`ast_…`) and pass the model id.

### OpenAI SDK

```js
import OpenAI from "openai";

const client = new OpenAI({
  baseURL: "https://gateway.agentsky.dev/v1",
  apiKey: process.env.AGENTSKY_API_KEY, // ast_…
});

const completion = await client.chat.completions.create({
  model: "deepseek/deepseek-v4-pro",
  messages: [{ role: "user", content: "Summarise this ticket." }],
});
console.log(completion.usage.cost); // USD, in the response
```

### Codex

```bash
export OPENAI_BASE_URL="https://gateway.agentsky.dev/v1"
export OPENAI_API_KEY="$AGENTSKY_API_KEY"

codex --model deepseek/deepseek-v4-pro
```

### OpenCode

```js
// opencode.json — then switch models any time with /models
{
  "$schema": "https://opencode.ai/config.json",
  "provider": {
    "agentsky": {
      "npm": "@ai-sdk/openai-compatible",
      "name": "AgentSky",
      "options": {
        "baseURL": "https://gateway.agentsky.dev/v1",
        "apiKey": "{env:AGENTSKY_API_KEY}"
      },
      "models": { "deepseek/deepseek-v4-pro": {} }
    }
  }
}
```

## Run it with an agent

- [Hermes + DeepSeek V4 Pro](https://agentsky.dev/agents/hermes/with/deepseek-v4-pro)
- [Claude Code + DeepSeek V4 Pro](https://agentsky.dev/agents/claude-code/with/deepseek-v4-pro)
- [Codex + DeepSeek V4 Pro](https://agentsky.dev/agents/codex/with/deepseek-v4-pro)
- [OpenClaw + DeepSeek V4 Pro](https://agentsky.dev/agents/openclaw/with/deepseek-v4-pro)
- [pi + DeepSeek V4 Pro](https://agentsky.dev/agents/pi/with/deepseek-v4-pro)
- [DeepSeek Harness + DeepSeek V4 Pro](https://agentsky.dev/agents/dsh/with/deepseek-v4-pro)
- [Kimi Code + DeepSeek V4 Pro](https://agentsky.dev/agents/kimi-code/with/deepseek-v4-pro)
- [OpenCode + DeepSeek V4 Pro](https://agentsky.dev/agents/opencode/with/deepseek-v4-pro)
- [Cline + DeepSeek V4 Pro](https://agentsky.dev/agents/cline/with/deepseek-v4-pro)
- [crow-cli + DeepSeek V4 Pro](https://agentsky.dev/agents/crow-cli/with/deepseek-v4-pro)
- [DeepAgents + DeepSeek V4 Pro](https://agentsky.dev/agents/deepagents/with/deepseek-v4-pro)
- [goose + DeepSeek V4 Pro](https://agentsky.dev/agents/goose/with/deepseek-v4-pro)
- [Grok Build + DeepSeek V4 Pro](https://agentsky.dev/agents/grok-build/with/deepseek-v4-pro)
- [Kilo + DeepSeek V4 Pro](https://agentsky.dev/agents/kilo/with/deepseek-v4-pro)
- [MiniMax Code + DeepSeek V4 Pro](https://agentsky.dev/agents/minimax-code/with/deepseek-v4-pro)
- [Mistral Vibe + DeepSeek V4 Pro](https://agentsky.dev/agents/mistral-vibe/with/deepseek-v4-pro)
- [Qwen Code + DeepSeek V4 Pro](https://agentsky.dev/agents/qwen-code/with/deepseek-v4-pro)
- [OpenHands + DeepSeek V4 Pro](https://agentsky.dev/agents/openhands/with/deepseek-v4-pro)

## Questions

### What is DeepSeek V4 Pro?

DeepSeek V4 Pro is a model from DeepSeek. The higher-capability V4 text model, with thinking mode on by default and a million-token window.

### How much does DeepSeek V4 Pro cost?

$0.924 / $2.772 per million tokens at the prices shown, for input and output respectively. The exact price of every call comes back in the response as usage.cost, so you never have to compute it.

### What is the context length of DeepSeek V4 Pro?

1M tokens, and it writes up to 384K tokens in one reply. These are the limits listed for this model in the current AgentSky catalog.

### Which clients can call DeepSeek V4 Pro?

Anything that speaks the OpenAI-compatible request format. Point the client's base URL at the gateway, use an AgentSky key, and pass deepseek/deepseek-v4-pro as the model.

### How do I name DeepSeek V4 Pro in a request?

The gateway accepts deepseek/deepseek-v4-pro, deepseek-v4-pro. Any of them resolves to the same model, so an existing integration usually needs no change to its model string.

### Can DeepSeek V4 Pro read images?

No image input is listed for this model. Its supported input types are text. For screenshots or photographs, choose a model that lists image input.

### What output can DeepSeek V4 Pro produce?

The listed output types are text. This model returns text. Producing image, video or audio files requires a suitable tool or another model; it is not included in text generation.

### How should I plan a long prompt for DeepSeek V4 Pro?

Plan your instructions, conversation history and source material around the 1M-token context limit, leaving room for the response. The listed response ceiling is 384K tokens. A larger window lets you supply more relevant material; it does not make unrelated content useful.

### What would 10,000 input tokens and 2,000 output tokens cost with DeepSeek V4 Pro?

At the displayed input and output rates, the model-token subtotal is $0.014784. This example assumes uncached input and excludes additional pricing categories, tool calls and agent runtime. Actual usage determines the final charge.

### How is cached input priced for DeepSeek V4 Pro?

The catalog lists $0.0308 per million cache-read tokens. Cache writes are listed at $0.924 per million tokens. Repeating a prompt does not by itself guarantee a cache hit; only usage reported as cached receives that rate.

### Does the DeepSeek V4 Pro token price cover an entire agent task?

No. The model price covers the model usage described above. A task can also incur tool and runtime charges, and several model calls may be needed to complete it. Compare total task usage when evaluating cost.

### How do I use DeepSeek V4 Pro on AgentSky?

The composer above preselects DeepSeek V4 Pro for a compatible agent. Enter a concrete task, select any tools or connectors it needs, and send. Sign in if prompted; paid execution requires available credits.