Name: BridgeML LLM – AI Text & Chat Generation | Free · $9/mo | API.market
Brand: bridgeml
SKU: api-bridgeml-llm
Availability: InStock
Rating: 5.00 (1 reviews)
Author: API Market

Shashank Agarwal

### 🚀🗂️🚀 BridgeML LLM API

The **BridgeML LLM API** offers robust language model interactions for generating human-like text, making it an invaluable tool for applications needing dynamic content generation, conversational agents, summarization, translation, and more. Built to support multiple high-performance language models, including Meta’s latest `llama` models, the API is optimized for developers looking to seamlessly integrate sophisticated language generation capabilities into their applications.

---

### Supported Models
BridgeML’s LLM API supports a range of language models, enabling developers to choose the model that best fits their requirements based on speed, context window size, and output quality.

### Models List

Below is the updated **BridgeML LLM API** model list, now including the "API Units per Token" column based on the provided multipliers.

| Model ID                           | Developer     | Context Window (tokens) | Max Tokens | Description                                                                 | API Units per Token |
|------------------------------------|---------------|--------------------------|------------|-----------------------------------------------------------------------------|----------------------|
| gemma-7b-it                        | Google        | 8,192                    | -          | Advanced LLM optimized for Italian.                                        | 2                    |
| gemma2-9b-it                       | Google        | 8,192                    | -          | Enhanced version of Gemma, optimized for Italian with larger parameters.   | 3                    |
| llama3-groq-70b-8192-tool-use-preview | Groq     | 8,192                    | -          | Tool-use enhanced model by Groq for multi-step generation tasks.           | 9                    |
| llama3-groq-8b-8192-tool-use-preview  | Groq     | 8,192                    | -          | Lighter Groq model optimized for quick responses in tool-use applications. | 3                    |
| llama-3.1-70b-versatile            | Meta          | 128k                     | 8,000      | High-parameter model for complex language understanding tasks.             | 8                    |
| llama-3.1-8b-instant               | Meta          | 128k                     | 8,000      | Quick response LLM optimized for conversational tasks.                     | 2                    |
| llama-3.2-11b-text-preview         | Meta          | 128k                     | 8,000      | Preview model for text-based tasks with 11 billion parameters.             | 4                    |
| llama-3.2-11b-vision-preview       | Meta          | 128k                     | 8,000      | Vision-enabled version for image and text processing.                      | 4                    |
| llama-3.2-1b-preview               | Meta          | 128k                     | 8,000      | Preview model with 1 billion parameters for lightweight applications.      | 1                    |
| llama-3.2-3b-preview               | Meta          | 128k                     | 8,000      | Mid-sized preview model for general-purpose language tasks.                | 2                    |
| llama-3.2-90b-text-preview         | Meta          | 128k                     | 8,000      | High-parameter text preview model with 90 billion parameters.              | 10                   |
| llama-3.2-90b-vision-preview       | Meta          | 128k                     | 8,000      | Vision-enabled version of the 90B model for complex image tasks.           | 10                   |
| llama-guard-3-8b                   | Meta          | 8,192                    | -          | Security-focused LLM optimized for privacy-aware applications.             | 3                    |
| llama3-70b-8192                    | Meta          | 8,192                    | -          | Versatile high-parameter model for comprehensive language tasks.           | 8                    |
| llama3-8b-8192                     | Meta          | 8,192                    | -          | High-speed model for general-purpose and conversational AI tasks.          | 2                    |
| mixtral-8x7b-32768                 | Mistral       | 32,768                   | -          | Large-context model optimized for long document processing and summaries.  | 4                    |


Each model’s API Units per Token rate reflects the unit multiplier, offering guidance on cost efficiency per token across various tasks.

**Example Model**: For this example, we use `llama-3.1-8b-instant`, a model known for its balance between speed and output quality.

---

### Key Features
1. **Conversational AI**: Supports `role-based` message exchange (e.g., `user`, `assistant`, `system`) for creating interactive conversations.
2. **High Flexibility with Model Selection**: Choose from multiple models based on complexity, token context window, and speed.
3. **Streaming Support**: Provides real-time, incremental responses ideal for dynamic user experiences.
4. **Fine-Tuned Output Control**: Configure `temperature`, `max_tokens`, `top_p`, and stopping conditions for precise output.

---

### Use Cases
- **Interactive Chatbots**: Build engaging, context-aware chatbots for customer support, e-commerce, and informational services.
- **Dynamic Content Generation**: Generate articles, stories, or marketing content that adapts based on user input or pre-set prompts.
- **Automated Summarization**: Provide brief summaries for lengthy documents or complex discussions.
- **Real-Time Translations**: Leverage multilingual capabilities to provide real-time translations within a conversation.
- **Education and Tutoring**: Use the API to create language-based tutoring tools, interactive learning modules, or answer-based educational apps.

---

### API Endpoints & Detailed Examples

#### 1. **Chat Completion API**

**Endpoint**: `POST https://api.magicapi.dev/api/v2/bridgeml/llm/llm/chat/completions`

**Description**: Creates a conversational completion for a series of messages, allowing users to engage in natural language conversations with the model.

**Example Request**:
```bash
curl -X 'POST' \
  'https://api.magicapi.dev/api/v2/bridgeml/llm/llm/chat/completions' \
  -H 'accept: application/json' \
  -H 'x-magicapi-key: YOUR_API_KEY' \
  -H 'Content-Type: application/json' \
  -d '{
  "messages": [
    {
      "role": "system",
      "content": "Write a nice long poem"
    },
    {
      "role": "user",
      "content": ""
    }
  ],
  "model": "llama-3.1-8b-instant",
  "temperature": 1,
  "max_tokens": 1024,
  "top_p": 1,
  "stream": false,
  "stop": "string"
}'
```

**Example Response**:
```json
{
    "created": 1731320092,
    "model": "llama-3.1-8b-instant",
    "choices": [
        {
            "index": 0,
            "message": {
                "role": "assistant",
                "content": "Here's a short poem:\n\nThe sunset fades to grey,\nThe stars begin their evening sway,\nA gentle breeze whispers low,\nAs night descends, and darkness grows.\n\nThe world is quiet, still and deep,\nThe moon's soft light begins to creep,\nA peaceful hush, a soothing sight,\nAs night's tranquility takes flight."
            },
            "logprobs": null,
            "finish_reason": "stop"
        }
    ]
}
```

---

#### 2. **Streaming Chat Completion API**

**Description**: Generates a completion for a set of messages, with token-by-token streaming for real-time applications. Streaming is beneficial for chatbots and other applications where immediate feedback improves user engagement.

**Example Streaming Request**:
```bash
curl -X 'POST' \
  'https://api.magicapi.dev/api/v2/bridgeml/llm/llm/chat/completions' \
  -H 'accept: application/json' \
  -H 'x-magicapi-key: YOUR_API_KEY' \
  -H 'Content-Type: application/json' \
  -d '{
  "messages": [
    {
      "role": "system",
      "content": "Write a nice long poem"
    },
    {
      "role": "user",
      "content": ""
    }
  ],
  "model": "llama-3.1-8b-instant",
  "temperature": 1,
  "max_tokens": 1024,
  "top_p": 1,
  "stream": true,
  "stop": "string"
}'
```

**Example Streaming Response** (Example Output):
```plaintext
data: {"created":1731320136,"model":"llama-3.1-8b-instant","choices":[{"index":0,"delta":{"role":"assistant","content":""},"logprobs":null,"finish_reason":null}]}
data: {"id":"chatcmpl-ad18175b-0887-40e4-abfd-ad0164dfa53a","object":"chat.completion.chunk","created":1731320136,"model":"llama-3.1-8b-instant","system_fingerprint":"fp_9cb648b966","choices":[{"index":0,"delta":{"role":"assistant","content":""},"logprobs":null,"finish_reason":null}]}

data: {"id":"chatcmpl-ad18175b-0887-40e4-abfd-ad0164dfa53a","object":"chat.completion.chunk","created":1731320136,"model":"llama-3.1-8b-instant","system_fingerprint":"fp_9cb648b966","choices":[{"

index":0,"delta":{"content":"In"},"logprobs":null,"finish_reason":null}]}
data: {"id":"chatcmpl-ad18175b-0887-40e4-abfd-ad0164dfa53a","object":"chat.completion.chunk","created":1731320136,"model":"llama-3.1-8b-instant","system_fingerprint":"fp_9cb648b966","choices":[{"index":0,"delta":{"content":" twilight"},"logprobs":null,"finish_reason":null}]}
data: {"id":"chatcmpl-ad18175b-0887-40e4-abfd-ad0164dfa53a","object":"chat.completion.chunk","created":1731320136,"model":"llama-3.1-8b-instant","system_fingerprint":"fp_9cb648b966","choices":[{"index":0,"delta":{"content":"'s"},"logprobs":null,"finish_reason":null}]}
```

---

### Parameters Explained

- **messages**: An array of message objects, with `role` and `content`. Roles include `system` (initial instructions), `user` (user queries), and `assistant` (model response).
- **model**: Specifies the model to use. For example, `llama-3.1-8b-instant`.
- **temperature**: Controls creativity. Higher values (e.g., 1) generate more diverse outputs.
- **max_tokens**: Limits the number of tokens in the output. Adjust based on response length requirements.
- **top_p**: Controls token probability sampling. Lower values make responses more deterministic.
- **stream**: When set to `true`, the API streams results token by token.
- **stop**: Specifies a stopping condition for the model output, like a specific character or word.

### Getting Started

To begin using the BridgeML LLM API:
- Register on [API.market](https://api.market) and obtain an API key.
- Use the key in the `x-magicapi-key` header for authorization.

---

### Pricing

Below is the pricing for the available AI models in the BridgeML LLM API. Each model is priced per million tokens, with the number of tokens provided per $1 based on current rates.

| **AI Model**                              | **Speed (Tokens/sec)** | **Token Price (Per Million Tokens)** | **Tokens per $1** |
|-------------------------------------------|------------------------|--------------------------------------|-------------------|
| **deepseek-r1-distill-qwen-32b**          | 140                    | $0.76                                | 1.32M             |
| **deepseek-r1-distill-llama-70b**         | 275                    | $1.09                                | 0.92M             |
| **deepseek-r1-distill-llama-70b-specdec** | 275                    | $1.09                                | 0.92M             |
| **qwen-2.5-32b**                          | 200                    | $0.76                                | 1.32M             |
| **Llama 3.2 1B (Preview) 8k**             | 3100                   | $0.044                               | 22.7M             |
| **Llama 3.2 3B (Preview) 8k**             | 1600                   | $0.066                               | 15.4M             |
| **Llama 3.1 70B Versatile 128k**          | 250                    | $0.869                               | 1.54M             |
| **Llama 3.1 8B Instant 128k**             | 750                    | $0.088                               | 11.36M            |
| **Llama 3 70B 8k**                        | 330                    | $0.869                               | 1.54M             |
| **Llama 3 8B 8k**                         | 1250                   | $0.088                               | 11.36M            |
| **Mixtral 8x7B Instruct 32k**             | 575                    | $0.264                               | 3.79M             |
| **Gemma 7B 8k Instruct**                  | 950                    | $0.077                               | 12.99M            |
| **Gemma 2 9B 8k**                         | 500                    | $0.22                                | 4.55M             |
| **Llama 3 Groq 70B Tool Use Preview 8k**  | 335                    | $0.979                               | 1.02M             |
| **Llama 3 Groq 8B Tool Use Preview 8k**   | 1250                   | $0.209                               | 4.77M             |
| **Llama Guard 3 8B 8k**                   | 765                    | $0.22                                | 4.55M             |


Each model’s speed and token count make it ideal for specific applications, from high-speed conversational AI to advanced content generation and analysis. Select the model that best fits your needs and start integrating powerful language generation into your application.


### API Units per Token

The **API Units per Token** measure provides a clear understanding of model usage costs. Each model has a unique multiplier that specifies the API units consumed per token. This multiplier enables developers to gauge the cost-effectiveness of each model relative to its capabilities.

For instance, a model with an API Unit multiplier of `2` will use two API units for each token generated, making it a budget-friendly option for applications prioritizing high-speed responses. Conversely, models with higher multipliers, such as `10`, may consume more API units per token but offer enhanced accuracy, broader context handling, or specialized features.

This metric helps developers make informed choices to optimize costs while meeting application requirements in areas such as conversation, content generation, or summarization. Selecting the appropriate model and unit rate supports efficient API usage and cost control across varied use cases.

### Accessing All Models via OpenAI SDK

This API enables access to a wide selection of advanced language models through the official OpenAI SDK, providing a seamless way to integrate various AI capabilities into your applications.

#### Quick Start with OpenAI SDK

The following example demonstrates how to connect to the BridgeML LLM API using the OpenAI SDK to access and interact with all available models.

```javascript
import OpenAI from 'openai';

const client = new OpenAI({
  apiKey: "YOUR_API_MARKET_API_KEY",
  baseURL: "https://api.magicapi.dev/api/v2/bridgeml/llm/llm/chat/completions", 
});

async function main() {
  const chatCompletion = await client.chat.completions.create({
    messages: [{ role: 'user', content: 'Say this is a test' }],
    model: 'llama-3.1-8b-instant',
  });
  console.log(chatCompletion);
}

main();
```

In the example above:
- **API Key**: Replace `"YOUR_API_MARKET_API_KEY"` with your actual API key.
- **Model**: Specify your desired model (e.g., `'llama-3.1-8b-instant'`) from the list below.

#### Models Available

Using this setup, you can access the following models:


- **deepseek-r1-distill-qwen-32b**
- **deepseek-r1-distill-llama-70b**
- **deepseek-r1-distill-llama-70b-specdec**
- **qwen-2.5-32b**
- **gemma-7b-it**
- **gemma2-9b-it**
- **llama3-groq-70b-8192-tool-use-preview**
- **llama3-groq-8b-8192-tool-use-preview**
- **llama-3.1-70b-versatile**
- **llama-3.1-8b-instant**
- **llama-3.2-11b-text-preview**
- **llama-3.2-11b-vision-preview**
- **llama-3.2-1b-preview**
- **llama-3.2-3b-preview**
- **llama-3.2-90b-text-preview**
- **llama-3.2-90b-vision-preview**
- **llama-guard-3-8b**
- **llama3-70b-8192**
- **llama3-8b-8192**
- **mixtral-8x7b-32768**

### Usage
Simply specify the model name in the `model` field of the `chat.completions.create` method to use any model from the list above. This flexibility enables you to switch between models based on your application requirements, such as speed, cost, or complexity.

This integration allows for a streamlined development experience with easy access to a wide array of powerful models, perfect for various use cases including text generation, chatbots, and advanced NLP tasks.

### Why Choose BridgeML’s LLM API?

With support for a wide variety of LLMs optimized for different use cases, BridgeML’s LLM API provides unmatched flexibility for developers. From chatbots to content generation, each feature is designed to enhance user experience through dynamic AI interactions, real-time response, and tailored model selection. Perfect for AI applications in education, business, entertainment, and customer service.

Generate text, build chatbots, and summarize content with 15+ LLMs, including Meta's Llama models. Free 100000 calls/mo, from $9/mo. Start free on API.market.

BridgeML LLM – AI Text & Chat Generation | Free · $9/mo | API.market

FREE