> ## Documentation Index
> Fetch the complete documentation index at: https://docs.venice.ai/llms.txt
> Use this file to discover all available pages before exploring further.

# Migrate from OpenAI

> Move OpenAI-compatible chat, embeddings, and image apps to Venice in minutes by swapping the base URL and API key — same SDK, more privacy.

Venice AI is a **drop-in replacement** for OpenAI. Same SDK, same code — just change two lines. Get privacy-first inference, uncensored models, and competitive pricing.

## The 2-Line Migration

### Python

```python theme={"system"}
# Before (OpenAI)
from openai import OpenAI
client = OpenAI()

# After (Venice) — change api_key and base_url
from openai import OpenAI
client = OpenAI(
    api_key="your-venice-api-key",          # ← Change 1
    base_url="https://api.venice.ai/api/v1"  # ← Change 2
)
```

### Node.js

```javascript theme={"system"}
// Before (OpenAI)
import OpenAI from 'openai';
const client = new OpenAI();

// After (Venice)
import OpenAI from 'openai';
const client = new OpenAI({
  apiKey: 'your-venice-api-key',
  baseURL: 'https://api.venice.ai/api/v1',
});
```

### cURL

```bash theme={"system"}
# Before
curl https://api.openai.com/v1/chat/completions ...

# After — just change the URL and key
curl https://api.venice.ai/api/v1/chat/completions ...
```

### Environment Variables

```bash theme={"system"}
# Before
OPENAI_API_KEY=sk-...
OPENAI_BASE_URL=https://api.openai.com/v1

# After
OPENAI_API_KEY=your-venice-api-key
OPENAI_BASE_URL=https://api.venice.ai/api/v1
```

<Tip>
  Many libraries and tools read `OPENAI_API_KEY` and `OPENAI_BASE_URL` automatically. Just updating these env vars may be all you need.
</Tip>

## Model Mapping

| OpenAI Model           | Venice Equivalent                              | Type       | Pricing (Input/Output per 1M) |
| ---------------------- | ---------------------------------------------- | ---------- | ----------------------------- |
| gpt-4o                 | `zai-org-glm-4.7` (Private)                    | Text       | $0.55 / $2.65                 |
| gpt-4o                 | `openai-gpt-52` (Anonymized)                   | Text       | $2.19 / $17.50                |
| gpt-4o-mini            | `qwen3-4b`                                     | Text       | $0.05 / $0.15                 |
| gpt-4-turbo            | `mistral-31-24b`                               | Text       | $0.50 / $2.00                 |
| o1 / o3                | `qwen3-235b-a22b-thinking-2507` (Private)      | Reasoning  | $0.45 / $3.50                 |
| o1 / o3                | `grok-41-fast` (Anonymized)                    | Reasoning  | $0.50 / $1.25                 |
| gpt-4-vision           | `mistral-31-24b` or `qwen3-vl-235b-a22b`       | Vision     | $0.50 / $2.00                 |
| text-embedding-3-small | `text-embedding-bge-m3`                        | Embeddings | $0.15 / $0.60                 |
| dall-e-3               | `qwen-image` (Private, \$0.01) or `flux-2-pro` | Image      | From \$0.01                   |
| whisper                | `nvidia/parakeet-tdt-0.6b-v3`                  | STT        | \$0.0001/sec                  |
| tts-1                  | `tts-kokoro`                                   | TTS        | \$3.50/1M chars               |

## Feature Compatibility

| Feature           | OpenAI | Venice | Notes                                         |
| ----------------- | ------ | ------ | --------------------------------------------- |
| Chat Completions  | ✅      | ✅      | Fully compatible                              |
| Streaming         | ✅      | ✅      | SSE format identical                          |
| Function Calling  | ✅      | ✅      | Same `tools` parameter                        |
| Structured Output | ✅      | ✅      | Same `response_format`                        |
| Vision            | ✅      | ✅      | Same content array format                     |
| Embeddings        | ✅      | ✅      | Same API                                      |
| Image Generation  | ✅      | ✅      | OpenAI-compatible via `/images/generations`\* |
| TTS               | ✅      | ✅      | Compatible                                    |
| STT               | ✅      | ✅      | Compatible                                    |
| Assistants API    | ✅      | ❌      | Use Characters or Minds instead               |
| Batch API         | ✅      | ❌      | Not yet available                             |
| Fine-tuning       | ✅      | ❌      | Not available                                 |

\*Venice also provides an OpenAI-compatible endpoint at `POST /images/generations` for easier migration from DALL-E. For Venice's native image API with additional options, see [Image Generate](/api-reference/endpoint/image/generate).

## Venice-Only Features

Venice offers capabilities OpenAI doesn't:

### 1. Built-in Web Search

```python theme={"system"}
response = client.chat.completions.create(
    model="venice-uncensored",
    messages=[{"role": "user", "content": "Latest AI news today"}],
    extra_body={
        "venice_parameters": {
            "enable_web_search": "auto"
        }
    }
)
```

### 2. Web Scraping

```python theme={"system"}
response = client.chat.completions.create(
    model="venice-uncensored",
    messages=[{"role": "user", "content": "Summarize https://example.com/article"}],
    extra_body={
        "venice_parameters": {
            "enable_web_scraping": True
        }
    }
)
```

### 3. Characters (AI Personas)

```python theme={"system"}
response = client.chat.completions.create(
    model="venice-uncensored",
    messages=[{"role": "user", "content": "Tell me about yourself"}],
    extra_body={
        "venice_parameters": {
            "character_slug": "venice-ai"
        }
    }
)
```

### 4. Uncensored Models

Venice's private models have no content filtering, making them suitable for:

* Creative writing without guardrails
* Security research and red teaming
* Honest analysis without refusal patterns
* Medical/legal information without disclaimers

### 5. Video Generation

```python theme={"system"}
# Queue a video generation job
import requests

response = requests.post(
    "https://api.venice.ai/api/v1/video/queue",
    headers={"Authorization": f"Bearer {api_key}", "Content-Type": "application/json"},
    json={
        "model": "wan-2.6-text-to-video",
        "prompt": "A serene lake at sunset with gentle waves",
        "resolution": "720p",
        "duration": 5,
    }
)
job_id = response.json()["id"]
```

## Why Migrate?

### Privacy

* **Zero data retention** on private models — your prompts are never stored
* **No training on your data** — ever
* OpenAI retains data for 30 days and may use it for safety research

### Cost

* Private models are **often cheaper** than OpenAI equivalents
* `qwen3-4b` at \$0.05/1M input is 10x cheaper than gpt-4o-mini
* `venice-uncensored` at $0.20/1M input vs gpt-4o at $2.50/1M

### Freedom

* **No content filtering** on uncensored models
* No account suspensions for controversial use cases
* Web3-native with crypto payment options
* DIEM staking for daily credits

### Model Diversity

* Access to models from multiple providers (Qwen, Llama, Mistral, Gemma, Claude, GPT, Grok, etc.)
* Switch between private and anonymized models per request
* New models added regularly

## Framework Migration

Most AI frameworks work with Venice by changing the base URL:

| Framework     | Change Required                     |
| ------------- | ----------------------------------- |
| LangChain     | `base_url` in `ChatOpenAI`          |
| Vercel AI SDK | `baseURL` in `createOpenAI`         |
| CrewAI        | `OPENAI_API_BASE` env var           |
| LlamaIndex    | `api_base` in `OpenAI`              |
| AutoGen       | `base_url` in config                |
| Haystack      | `api_base_url` in `OpenAIGenerator` |
| Claude Code   | `--api-base` flag or env var        |
| Cursor        | Custom API endpoint in settings     |
| Continue.dev  | `apiBase` in config.json            |

<Card title="Get Your API Key" icon="key" href="https://venice.ai/settings/api">
  Generate a Venice API key and start migrating in minutes
</Card>