# Getting Started with Humiris AI in Cline Source: https://docs.humiris.ai/Cline Save 2 hours per developer every single day in vscode with cline ### **Step 1: Install Cline in Visual Studio Code** 1. Open **VS Code**. 2. Go to the **Extensions** tab (or press Ctrl+P and type ext install cline.cline). 3. Install the extension called **Cline**.

Once installed, click the **Cline icon** in the sidebar to launch it. **Step 2: Get Your Humiris API Key** 1. Go to[ ](https://www.humiris.ai)[humiris.ai](http://humiris.ai) and create an account. 2. After logging in, go to the **API Keys** section. 3. Click **“Create API Key”** to generate a new key. 4. Copy the key for later.

### **Step 3: Configure Cline with Humiris** 1. Inside Cline, click the **Settings (⚙️)** button. 2. Under **API Provider**, select **OpenAI Compatible**. 3. Fill out the fields as follows: * **Base URL**: [https://api.humiris.ai/api/openai/v1](https://api.humiris.ai/api/openai/v1) * **API Key**: Paste the API key from your Humiris account. * **Model ID**: Choose one of the supported Humiris models (e.g. codiris-v1-preview, codiris-v1-mini,daddy1,baby1, mommy1-adv, etc.) Click **Save**.

### **Step 4: Use Humiris Models in Cline** Once setup is complete: * You can now write prompts or coding requests directly into Cline. * Cline will use the selected **Humiris model** to help with code generation, explanations, and other dev tasks.

# Getting Started with Humiris AI in Cursor Source: https://docs.humiris.ai/Cursor Learn how to add Humiris models in cursor ### **Getting Started with Humiris AI in Cursor** #### **STEP 1: Installing Cursor** 1. Download and install Cursor from[https://www.cursor.com/downloads](https://www.cursor.com/downloads) 2. Open the **Cursor app**. #### **STEP 2: Adding Humiris AI Models** 1. In the **top-left corner**, click the **“Cursor” tab** in the menu bar. In the dropdown, click **“Settings”**, then choose **“Cursor Settings.”** 2. Once you're in Cursor Settings, go to the **“Models”** tab. 3. In the **Models** tab, first **unselect all currently selected models.**

4. Then, click the **➕ (plus)** button to **add the Humiris models**. You can add **one**, **some**, or **all** of the following by typing their name: * **baby1** – a small, fast, general-purpose model * **baby1-adv** – an advanced version of baby1 with improved reasoning * **mommy1** – larger, more capable, and creative than baby1 * **mommy1-adv** – an advanced version of mommy1 with stronger problem-solving * **daddy1** – the most capable model, ideal for complex tasks * **daddy1-adv** – an advanced version of daddy1, offering the highest performance * **codiris-v1-preview** - our first multimodal, post-trained engineered specifically for complex code workflows. * \*\*codiris-v1-mini - \*\*Cost-efficient, post-trained built for coding

5. In the **API Keys** section of the Models tab: * Turn OFF the following providers: Anthropic, Google, Azure, and AWS * **Only OpenAI should be toggled ON (green).** #### **STEP 3: Create a Humiris Account and Get Your API Key** 1. Go to[ ](https://humiris.ai)[humiris.ai](http://humiris.ai) and **create an account**. 1. When you add your credit card, we’ll automatically give you **\$5 in free credits** to get started. 2. Once logged in, go to **API Keys** and click on \*\*“Create API Key” \*\*to generate a new API key. 3. **Copy the API key**.

#### **STEP 4: Complete Setup in Cursor** 1. Back in Cursor, paste your **Humiris API key** in the **OpenAI API Key** field.\ *(Don’t click “Verify” yet — there’s one more step.)* 2. Scroll down to the field labeled **“Override OpenAI Base URL”** and enter [https://api.humiris.ai/api/openai/v1](https://api.humiris.ai/api/openai/v1) 3. Click **Save**. 4. Now click **Verify** to finish setup.

\ **STEP 5: Use Humiris Models in Cursor** **Cursor support only now the ask mode of humiris models.** [**If you have some questions send message to contact@humiris.ai**](mailto:contact@humiris.ai)

# Hmodels OpenAI Compatible Source: https://docs.humiris.ai/Hmodel-OpenAiCompatible Integration guide for Hmodel OpenAI Compatibility with Python and JavaScript. # Introduction Our platform allows you to integrate our [Hmodels](https://docs.humiris.ai/Hmodels) (**baby1**, **daddy1**, **mommy1**) into your projects using OpenAI Compatibility. You can interact with these models using the official `openai` package. ## Generate an API Key Before you start, you need to generate an API key from our platform. To do so, visit: 👉 [Generate an API Key](https://console.humiris.ai/api-keys) ## Base URL All requests must be sent to: ``` https://api.humiris.ai/api/openai/v1 ``` ## Installing the OpenAI Package If you haven't installed it yet, add the OpenAI package to your project: **Python:** ```bash pip install openai ``` **JavaScript (Node.js):** ```bash npm install openai ``` *** # Using OpenAI Compatibility ## Sending a Request to a Model Here is an example of how to interact with the **baby1 basic** model using Python and JavaScript. Find an verview of all available AI models on Hmodels here 👉 [Hmodels](https://docs.humiris.ai/Hmodels). ```python # Python Example import openai openai.api_key = "your_api_key" openai.api_base = "https://api.humiris.ai/api/openai/v1" response = openai.ChatCompletion.create( model="codiris-v1-preview", messages=[ {"role": "system", "content": "You are a helpful assistant."}, {"role": "user", "content": "Explain the theory of relativity."} ] ) print(response["choices"][0]["message"]["content"]) ``` ```javascript // JavaScript Example import OpenAI from "openai"; const openai = new OpenAI({ apiKey: "your_api_key", baseURL: "https://api.humiris.ai/api/openai/v1", }); async function run() { const response = await openai.chat.completions.create({ model: "daddy1", messages: [ { role: "system", content: "You are a helpful assistant." }, { role: "user", content: "Explain the theory of relativity." }, ], }); console.log(response.choices[0].message.content); } run(); ``` *** ## Other Useful Parameters The API supports several options to customize the model's responses. | Parameter | Description | | ------------- | ---------------------------------------------------------------------- | | `model` | Name of the model used (`daddy1` or `daddy1-adv`). | | `messages` | List of messages to structure the conversation. | | `temperature` | Controls response creativity (0 = deterministic, 1 = highly creative). | | `max_tokens` | Maximum number of tokens in the response. | **Example with Advanced Parameters:** ```python # Python Example import os import openai openai.api_key = "your_api_key" openai.api_base = "https://api.humiris.ai/api/openai/v1" response = openai.ChatCompletion.create( model="daddy1", messages=[ {"role": "system", "content": "You are an expert in quantum physics."}, {"role": "user", "content": "Explain the tunnel effect."} ], temperature=0.7, max_tokens=200 ) ``` ```javascript // JavaScript Example import OpenAI from "openai"; const openai = new OpenAI({ apiKey: "your_api_key", baseURL: "https://api.humiris.ai/api/openai/v1", }); const response = await openai.chat.completions.create({ model: "daddy1", messages: [ { role: "system", content: "You are an expert in quantum physics." }, { role: "user", content: "Explain the tunnel effect." }, ], temperature: 0.7, max_tokens: 200, }); ``` *** # Hmodels Source: https://docs.humiris.ai/Hmodels Overview of all available AI models on Humiris platform. # Humiris AI Models Our platform offers multiple AI models categorized into **Basic** and **Advanced** versions. These models are designed for different use cases, providing flexibility and power based on your needs. ## Model Categories * **Basic Models**: Standard AI models optimized for general usage. * **Advanced Models**: Enhanced versions with improved capabilities and deeper contextual understanding. ## Available Models ### Basic Models | Model Name | Description | Pricing (per million tokens) | | ---------- | ------------------------------------------------------------------------------------------------------------------------------------------------ | ---------------------------- | | `baby1` | Optimized for cost-speed efficiency, making it ideal for rapid inference and real-time applications. | Input: $0.5, Output: $ 1 | | `mommy1` | Specializes in contextual reasoning, problem decomposition, and structured execution of complex workflows. | Input: $7, Output: $ 15 | | `daddy1` | Offers broad knowledge and deep contextual understanding, significantly reducing hallucinations and enhancing reliability across diverse topics. | Input: $3, Output: $ 15 | ### Advanced Models | Model Name | Description | Pricing (per million tokens) | | ------------ | ------------------------------------------------------------------------------------------------------------------------------------------------------------------------- | ---------------------------- | | `baby1-adv` | A cost-effective, high-speed reasoning model designed for optimal performance. | Input: $2, Output: $ 5 | | `mommy1-adv` | Balances accuracy and efficiency, making it versatile for structured reasoning tasks. | Input: $5, Output: $ 20 | | `daddy1-adv` | Humiris’ most advanced and intelligent AI model for STEM reasoning and complex problem-solving. Designed to push the boundaries of math, coding, and scientific analysis. | Input: $10, Output: $ 30 | ### **Vertical Models** Access vertical-specialized models tailored for code generation, healthcare, finance, and more ready-to-use via API. | **Model** | **Context Window (Tokens** | **Max Completion Token** | **Description** | Pricing (per million tokens) | | :--------------------- | :------------------------- | :----------------------- | :-------------------------------------------------------------------------------------------------- | ---------------------------- | | **codiris-v1-preview** | **1M** | **50,000** | **Codiris**, our first multimodal, post-trained engineered specifically for complex code workflows. | Input: \$10, Output: \$30 | | **codiris-v1-mini** | **1M** | **50,000** | Cost-efficient, post-trained built for coding | Input: \$3, Output: \$15 | ## Choosing the Right Model * Use **Basic Models** for standard AI-powered features requiring efficiency and speed. * Choose **Advanced Models** when you need deeper analysis, better contextual understanding, or more accurate responses. * Choose \*\*Vertical model \*\*for coding. (Other models for domain like Fintech, Healthtech will be available soon) # Node.js Source: https://docs.humiris.ai/Nodejs Humiris AI is a library that allows developers to seamlessly integrate Mixture of AI Basic and Advanced into their applications. This library is designed to simplify the process of adding Mixture of AI features and functionalities, making it accessible for developers of all levels **Prerequisite**: Please install Node.js (version 19 or higher) before proceeding. **Install Humiris** Open your terminal or command prompt and run the following command: ```bash npm npm install humiris-ai ``` ```bash yarn yarn add humiris-ai ``` **Prerequisite**: At this step you will need your Moai Apikey to continue.{" "} How to get my Moai Apikey ? **Importing the Library** After installation, you can import the Humiris library into your project: ```python import { BasicMoAi, AdvancedMoAi } from 'humiris-ai'; ``` **Basic Usage** ```typescript import { BasicMoAi } from "humiris-ai"; const moai = new BasicMoAi("your_api_key"); const payload = { mixTuningId: "exampleId", messages: [ { role: "system", content: "You are a helpful assistant that answers questions about technology.", }, { role: "user", content: "What is the diference between a microprocessor and a central processing unit?", }, ], temperature: 0.7, maxTokens: 150, }; moai .useBasicMixtuning(payload) .then((response) => { console.log("Response from AI:", response); }) .catch((error) => { console.error("Error:", error); }); ``` **Advanced Usage** ```typescript import { AdvancedMoAi } from "humiris-ai"; const advancedMoai = new AdvancedMoAi("your_api_key"); const advancedPayload = { mixTuningId: "advancedExampleId", messages: [ { role: "system", content: "You are a helpful assistant that answers questions about technology.", }, { role: "user", content: "What is the diference between a microprocessor and a central processing unit?", }, ], temperature: 0.9, maxTokens: 250, }; advancedMoai .useAdvancedMixtuning(advancedPayload) .then((response) => { console.log("Advanced response from AI:", response); }) .catch((error) => { console.error("Advanced error:", error); }); ``` **Support** For any questions or issues, please contact Hilario Houmey at [h.hilario@humiris.ai](mailto:h.hilario@humiris.ai). # OpenAi-Compatible Source: https://docs.humiris.ai/OpenAiCompatible Welcome to the **OpenAI-Compatible API** documentation. This guide will help developers integrate with our API seamlessly. It replicates OpenAI's functionality while providing additional features like `mixTuningId` for enhanced customization. # Python Source: https://docs.humiris.ai/Python Humiris AI is a library that allows developers to seamlessly integrate Mixture of AI Basic and Advanced into their applications. This library is designed to simplify the process of adding Mixture of AI features and functionalities, making it accessible for developers of all levels. **Prerequisite**: To install the Humiris AI library, you can use pip, the Python package manager. Ensure you have Python 3.6 or later installed on your system. **Install Humiris** Open your terminal or command prompt and run the following command: ```bash pip pip install humiris_ai ``` **Prerequisite**: At this step you will need your Mixture Apikey to continue.{" "} How to get my Moai Apikey ? **Importing the Library** After installation, you can import the Humiris library into your Python project: ```python import humiris_ai ``` **Basic Usage** ```python from humiris_ai import BasicMoAi, MessageType, BasicPayload # Initialize the MoAi with your API key moai = BasicMoAi(api_key='your_api_key') # Create a payload for the AI service messages = [ MessageType(role='system', content='You are a helpful assistant that answers questions about technology.'), MessageType(role='user', content='What is the diference between a microprocessor and a central processing unit?') ] payload = BasicPayload( mix_tuning_id='exampleId', messages=messages, temperature=0.7, max_tokens=150 ) # Call the AI service response = moai.use_basic_mixtuning(payload) print(response) ``` **Advanced Usage** ```python from humiris_ai import AdvancedMoAi, MessageType, AdvancedPayload # Initialize the MoAi with your API key moai = AdvancedMoAi(api_key='your_api_key') # Create a payload for the AI service messages = [ MessageType(role='system', content='You are a helpful assistant that answers questions about technology.'), MessageType(role='user', content='What is the diference between a microprocessor and a central processing unit?') ] payload = AdvancedPayload( advanced_mix_tuning_id='exampleId', messages=messages, temperature=0.7, max_tokens=150 ) # Call the AI service response = moai.use_advanced_mixtuning(payload) print(response) ``` **Support** For any questions or issues, please contact Hilario Houmey at [h.hilario@humiris.ai](mailto:h.hilario@humiris.ai). # Welcome to Humiris AI Documentation Source: https://docs.humiris.ai/Welcome Here you'll get an overview of all the amazing features Humiris offers to help you mix all LLMs in 1 model. Humiris API provides a simple interface to state-of-the-art AI [models](https://platform.openai.com/docs/models) for text generation, natural language processing, reasoning and more… Logoblanccopie Pn

### Key Features ### **Frontier Models** Unlock the full power of AI with our first Mix-Models family seamlessly combining top-tier models.Unlike conventional AI, Humiris mix-models function as a self-improving, autonomous brain network, capable of advanced reasoning, real-world interaction, and strategic decision-making. | **Model** | **Context Window (Tokens)** | **Max Completion Token** | **Description** | | :------------- | :-------------------------- | :----------------------- | :------------------------------------------------------------------------------------------------------------------------------------------------------------------------ | | **Baby1** | **128,000** | **4,096** | Optimized for cost-speed efficiency, making it ideal for rapid inference and real-time applications. | | **Baby1-adv** | **128,000** | **4,096** | A cost-effective, high-speed reasoning model designed for optimal performance. | | **Mommy1** | **128,000** | **16000** | Specializes in contextual reasoning, problem decomposition, and structured execution of complex workflows. | | **Mommy1-adv** | **128,000** | **16000** | Balances accuracy and efficiency, making it versatile for structured reasoning tasks. | | **Daddy1** | **200,000** | **16000** | Offers broad knowledge and deep contextual understanding, significantly reducing hallucinations and enhancing reliability across diverse topics. | | **Daddy1-adv** | **200,000** | **16000** | Humiris’ most advanced and intelligent AI model for STEM reasoning and complex problem-solving. Designed to push the boundaries of math, coding, and scientific analysis. | ### **Vertical Models** Access vertical-specialized models tailored for code generation, healthcare, finance, and more ready-to-use via API. | **Model** | **Context Window (Tokens** | **Max Completion Token** | **Description** | | :--------------------- | :------------------------- | :----------------------- | :-------------------------------------------------------------------------------------------------- | | **codiris-v1-preview** | **1M** | **50,000** | **Codiris**, our first multimodal, post-trained engineered specifically for complex code workflows. | | **codiris-v1-mini** | **1M** | **50,000** | Cost-efficient, post-trained built for coding | # Quickstart Mixture of AI Source: https://docs.humiris.ai/quickstart Start building your own mix-model ## 1. Register for an account Go to [Humiris](https://www.platform.humiris.ai) and sign up to create an account.

## 2. Create your first project Go to [Humiris](https://www.platform.humiris.ai) and sign up to create an account. Before we start make sure that you [create an account here](https://platform-moai-git-main-joe-ai.vercel.app/). After creating your account you will have a default project and workspace. Your project is linked to your workspace. All your data (API Keys, Models, etc.) are organized according to your workspace. This allows you to manage your projects independently and add collaborators.

Upon creating your account, you will be added to your default workspace and project. After that, you can create your own workspaces and projects.

## 3. Create your API key Create it in the "API Key" section by clicking on the "Create New Key" button.

After that, we will generate a secure key for you that will help you connect to our platform with complete security.

We're all set. We can now dive deeper. ## 4. Add your LLMs providers API Keys Obtain API keys from your chosen LLM providers (e.g., OpenAI, Anthropic). *If you don’t have an account on their platform, you must sign up and create API keys.* **Copy the Key**: After generating, copy the API key to use in your Humiris configuration.

## 5. Mix your models Choose the type of mixture you want to create: Basic or Advanced. The Basic option optimizes for preferences like speed, cost, energy efficiency, quality, and privacy. The Advanced option allows you to mix models based on a reasoning architecture. Select the models you wish and mix them using tuning parameters until you achieve the perfect mix-model Basic mixture

Advanced mixture