> ## Documentation Index
> Fetch the complete documentation index at: https://mintlify.com/cheahjs/free-llm-api-resources/llms.txt
> Use this file to discover all available pages before exploring further.

# Hyperbolic

> Get $1 in free credits to access DeepSeek, Llama, Qwen, and other models

## Overview

Hyperbolic provides \$1 in trial credits to new users, giving you access to a curated selection of powerful open-source models including DeepSeek V3, Llama 3.1 405B, and Qwen models.

<CardGroup cols={2}>
  <Card title="Trial Credits" icon="dollar-sign">
    \$1 in free credits
  </Card>

  <Card title="Duration" icon="clock">
    No expiration specified
  </Card>
</CardGroup>

## Available Models

Hyperbolic offers access to high-performance models optimized for their infrastructure:

### Language Models

<table>
  <thead>
    <tr>
      <th>Model</th>
      <th>Description</th>
    </tr>
  </thead>

  <tbody>
    <tr>
      <td>DeepSeek V3</td>
      <td>Latest DeepSeek model with strong reasoning</td>
    </tr>

    <tr>
      <td>DeepSeek V3 0324</td>
      <td>Specific checkpoint version</td>
    </tr>

    <tr>
      <td>deepseek-ai/deepseek-r1-0528</td>
      <td>DeepSeek reasoning model</td>
    </tr>

    <tr>
      <td>Llama 3.1 405B Base</td>
      <td>Largest Llama base model</td>
    </tr>

    <tr>
      <td>Llama 3.1 405B Instruct</td>
      <td>Instruction-tuned 405B model</td>
    </tr>

    <tr>
      <td>Llama 3.1 70B Instruct</td>
      <td>Efficient 70B instruction model</td>
    </tr>

    <tr>
      <td>Llama 3.1 8B Instruct</td>
      <td>Fast 8B instruction model</td>
    </tr>

    <tr>
      <td>Llama 3.2 3B Instruct</td>
      <td>Compact instruction model</td>
    </tr>

    <tr>
      <td>Llama 3.3 70B Instruct</td>
      <td>Latest 70B instruction model</td>
    </tr>

    <tr>
      <td>Qwen QwQ 32B</td>
      <td>Reasoning-focused model</td>
    </tr>

    <tr>
      <td>Qwen2.5 72B Instruct</td>
      <td>High-performance Qwen model</td>
    </tr>

    <tr>
      <td>Qwen2.5 Coder 32B Instruct</td>
      <td>Specialized coding model</td>
    </tr>

    <tr>
      <td>qwen/qwen3-235b-a22b</td>
      <td>Latest Qwen 3 model</td>
    </tr>

    <tr>
      <td>qwen/qwen3-235b-a22b-instruct-2507</td>
      <td>Instruction-tuned Qwen 3</td>
    </tr>

    <tr>
      <td>qwen/qwen3-coder-480b-a35b-instruct</td>
      <td>Large coding model</td>
    </tr>

    <tr>
      <td>qwen/qwen3-next-80b-a3b-instruct</td>
      <td>Next-gen Qwen model</td>
    </tr>

    <tr>
      <td>qwen/qwen3-next-80b-a3b-thinking</td>
      <td>Reasoning variant</td>
    </tr>

    <tr>
      <td>openai/gpt-oss-120b</td>
      <td>Open-source GPT model</td>
    </tr>

    <tr>
      <td>openai/gpt-oss-120b-turbo</td>
      <td>Optimized GPT variant</td>
    </tr>

    <tr>
      <td>openai/gpt-oss-20b</td>
      <td>Smaller open GPT model</td>
    </tr>
  </tbody>
</table>

### Vision Models

* **Pixtral 12B (2409)**: Multimodal vision-language model
* **Qwen2.5 VL 72B Instruct**: Large vision-language model
* **Qwen2.5 VL 7B Instruct**: Efficient vision-language model

## Getting Started

### 1. Sign Up

Visit [app.hyperbolic.xyz](https://app.hyperbolic.xyz/) and create a free account to receive your \$1 credit.

### 2. Get Your API Key

Navigate to your dashboard to generate an API key.

### 3. Make API Calls

```python theme={null}
import openai

client = openai.OpenAI(
    api_key="YOUR_HYPERBOLIC_API_KEY",
    base_url="https://api.hyperbolic.xyz/v1"
)

response = client.chat.completions.create(
    model="meta-llama/Meta-Llama-3.1-405B-Instruct",
    messages=[{
        "role": "user",
        "content": "What is the capital of France?"
    }]
)

print(response.choices[0].message.content)
```

```javascript theme={null}
import OpenAI from 'openai';

const client = new OpenAI({
  apiKey: process.env.HYPERBOLIC_API_KEY,
  baseURL: 'https://api.hyperbolic.xyz/v1'
});

const response = await client.chat.completions.create({
  model: 'meta-llama/Meta-Llama-3.1-405B-Instruct',
  messages: [{
    role: 'user',
    content: 'What is the capital of France?'
  }]
});

console.log(response.choices[0].message.content);
```

```bash theme={null}
curl https://api.hyperbolic.xyz/v1/chat/completions \
  -H "Authorization: Bearer YOUR_HYPERBOLIC_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "meta-llama/Meta-Llama-3.1-405B-Instruct",
    "messages": [{
      "role": "user",
      "content": "What is the capital of France?"
    }]
  }'
```

## Model Highlights

<CardGroup cols={2}>
  <Card title="DeepSeek V3" icon="brain">
    State-of-the-art reasoning and coding capabilities
  </Card>

  <Card title="Llama 3.1 405B" icon="star">
    One of the largest and most capable open models
  </Card>

  <Card title="Qwen 3" icon="language">
    Excellent multilingual and coding performance
  </Card>

  <Card title="Vision Models" icon="image">
    Multimodal understanding with Qwen VL and Pixtral
  </Card>
</CardGroup>

## Use Cases

* **Advanced Reasoning**: Use DeepSeek or Qwen QwQ for complex problems
* **Code Generation**: Leverage specialized coding models
* **Vision Tasks**: Process images with multimodal models
* **Large Context**: Use 405B models for complex, long-context tasks
* **Multilingual**: Qwen models excel at Chinese and other languages

## API Compatibility

<Note>
  Hyperbolic uses an **OpenAI-compatible API**, making it easy to integrate with existing code and tools.
</Note>

## Resources

<CardGroup cols={2}>
  <Card title="Hyperbolic Platform" icon="link" href="https://app.hyperbolic.xyz/">
    Access the platform
  </Card>

  <Card title="Dashboard" icon="gauge">
    View usage and credits
  </Card>
</CardGroup>

<Info>
  Start with smaller models (8B-32B) to maximize your \$1 credit, then test larger models for specific use cases.
</Info>