Loading...
Loading...
Loading...
Loading...
Loading...
Loading...
Loading...
Loading...
Loading...
Loading...
Loading...
Loading...
Loading...
Loading...
Loading...
Loading...
Loading...
Loading...
Loading...
Loading...
Loading...
Loading...
Loading...
Loading...
Loading...
Loading...
Loading...
Loading...
Loading...
Loading...
Loading...
Loading...
Loading...
Loading...
Loading...
Loading...
Loading...
Loading...
Loading...
Loading...
Loading...
Loading...
Accessing ChatNGO is simple and designed to accommodate users from various organizations or roles. Whether you are part of multiple nonprofits or a single team, you can get started quickly and easily by following the steps below.
Step 1: Accept Your Invitation to Join ChatNGO You will receive an email invitation from your organization's administrator. Click the link to accept the invitation.
Step 2: Create Your Account Enter your name, email address, and password. Verify your email by following the instructions in the confirmation email.
Step 3: Log In and Select Your Workspace Go to the ChatNGO platform (https://team.chat.ngo) or click the login button on the website. Enter your credentials and select your workspace if applicable.
ChatNGO workspaces are designed to keep your projects, teams, and organizations organized and secure. Each workspace is a private environment where you and your team can collaborate on AI chats, share prompts, and manage workflows.
A workspace in ChatNGO is a dedicated space for a specific organization or team. Workspaces allow users to keep their conversations, images, and custom AI models separate from other teams or organizations they may be a part of.
Each workspace contains:
Private and Shared Folders for organizing chats, files, and prompts.
Custom Instructions and Personas to tailor AI outputs for your team.
Permissions and Roles that manage who has access to different sections within the workspace.
AI Model Settings that determine which AI models are available to your team.
Multiple Workspaces: Users can be a part of multiple workspaces if they collaborate with different nonprofits or collaborative projects.
Team Collaboration: Workspaces enable your team to chat, share prompts, and manage AI models collaboratively.
Customized AI Models: Each workspace can have its own set of AI models, specific to the needs of the team.
Secure Environment
You may have access to multiple workspaces in certain cases:
Different Organizations or Roles: If you are associated with more than one nonprofit, for example, as a volunteer for one and a board member for another, you may be invited to join separate workspaces for each organization.
Collaborative Projects: Sometimes, multiple nonprofits work together on a shared project. In such cases, a dedicated workspace may be created for the project, allowing team members from each organization to collaborate in one space.
In each scenario, workspace admins will invite you, and you can easily switch between your workspaces to manage your work across different organizations or projects.
ChatNGO is offered completely free of charge to nonprofit organizations. There are no monthly subscription fees, no user fees, and no additional charges from ChatNGO. This enables nonprofits to leverage AI tools without financial barriers directly through our platform.
However, to access AI functionality within ChatNGO, your organization will need to set up API keys from third-party AI providers such as OpenAI (for GPT models) or Anthropic (for Claude models). Here is a breakdown of what to expect regarding costs and responsibilities:
Nonprofits are responsible for covering the cost of AI usage directly with the chosen AI providers (OpenAI, Anthropic, etc.).
Each provider has its own pricing structure for different models. For example, GPT-4 and Claude models have varying rates depending on their capabilities and token usage.
We recommend that you carefully review the usage fees for each model and provider. Setting up usage limits with each provider is advised to manage and control your budget.
While the costs will vary based on usage and selected models, we generally recommend budgeting around $5 per user per month. This estimate provides a guideline but may differ depending on how often and intensively users interact with the AI models.
For detailed instructions on how to create and link an API key, please refer to our documentation on . This guide will walk you through the steps needed to connect your workspace to these AI services, enabling your team to access AI-powered features on ChatNGO.
Each model’s usage can add up quickly depending on the volume of interactions. To avoid unexpected costs, regularly monitor your usage and set limits as necessary through your chosen AI provider’s dashboard.
Both OpenAI and Anthropic provide options to control and monitor API usage, allowing you to set a cap on expenses. We recommend reviewing these tools to help manage costs effectively.
By covering only the direct AI usage costs with third-party providers, ChatNGO provides a cost-effective solution for nonprofits, ensuring they can access the latest in AI technology without incurring platform fees.
Prompt Library: Access ready-made prompts or create your own, tailored to your organization’s needs.
User-Friendly Interface: Navigate easily and utilize prompt libraries, making AI accessible even for non-technical users.
Security and Privacy: Protect your data and models with strict privacy controls and workspace-specific permissions.
Real-Time Reporting: Track AI usage and its impact through detailed reporting to optimize performance.
ChatNGO is powered by Team-GPT, an enterprise-level AI platform designed to enhance productivity, collaboration, and automation through advanced AI models. While users will see Team-GPT branding throughout the platform, ChatNGO is a licensed, customized version specifically tailored to the needs of NGOs and nonprofit organizations. Below, we explain how Team-GPT works behind the scenes to support ChatNGO's capabilities.
Team-GPT provides the core infrastructure, AI models, and features that make ChatNGO possible. It handles everything from platform updates to integrating AI models like GPT-4o and Sonnet 3.5, ensuring the system remains powerful and adaptable. Key components of Team-GPT's role include:
Platform Updates: Ensures ChatNGO always runs on the latest technology.
AI Model Integration: Supports the seamless integration of models like GPT-4o and Sonnet 3.5.
Infrastructure Management: Provides the robust backend needed to power advanced features.
While Team-GPT is a general enterprise solution, ChatNGO is customized to meet the unique needs of NGOs. We have tailored its features, privacy controls, and user interfaces to align with nonprofit workflows, ensuring a seamless user experience. Specific customizations include:
Privacy Controls: Enhanced privacy measures to suit nonprofit environments.
Custom Features: Tools and workflows specifically designed for nonprofit users.
Team-GPT is responsible for maintaining the platform and keeping it updated with the latest features. Although Team-GPT branding is visible, ChatNGO operates on dedicated servers, providing enterprise-level software with nonprofit-specific features and security.
At ChatNGO, we prioritize your security and privacy. Although powered by Team-GPT, ChatNGO runs on dedicated servers to ensure your data remains private and secure.
Data Privacy and Encryption: All data shared or uploaded in ChatNGO is protected with advanced encryption (AES-256) and remains private to your organization. Your data is never used for training purposes.
No Access to User Data: Team-GPT does not access user data, including chat history, emails, or personal details. Your data remains fully isolated within your workspace.
GDPR Compliance: Team-GPT complies with GDPR regulations, ensuring that EU-based users' privacy rights are protected.
Team-GPT is working toward important security certifications and compliance standards to ensure robust data protection:
ISO/IEC 27001: A globally recognized standard for managing information security.
SOC 2 Compliance: Focuses on data security, availability, and confidentiality.
The Chatting with ChatNGO section provides all the tools and instructions needed to effectively communicate with ChatNGO. This section guides you through starting new conversations, selecting language models, and utilizing the additional features that enhance your experience. From attaching files to customizing your instructions, this section ensures you get the most out of ChatNGO’s conversational abilities.
Starting a New Chat: Learn how to initiate a conversation with ChatNGO, whether for personal use or collaborative purposes.
Selecting a Language Model: Discover how to choose between different AI models like GPT-4o, GPT-4o mini, or Claude 3.5 Sonnet, depending on your specific needs.
Attaching Images or Files: Explore how to upload and attach images to your chat conversations to enhance your communication with ChatNGO.
Prompt Library: Access a library of pre-built prompts that you can use to streamline conversations and gather insightful information from ChatNGO.
Custom Instructions: Personalize your interaction by adding specific instructions for ChatNGO to tailor responses to your needs.
Generating Images: Learn how to use ChatNGO’s image generation capabilities to create visuals directly in the chat.
The Workspace Settings section allows admins and managers to configure key elements of their workspace, ensuring it is tailored to meet the needs of the organization. From managing members to configuring AI models, this section provides all the tools necessary for maintaining and optimizing a workspace.
From your workspace, click on Your Workspace Name on the top menu.
Select Workspace Settings from the dropdown options.
To access and manage your profile in ChatNGO, follow these steps:
Click on Profile Icon: Located at the bottom left of the ChatNGO platform, you will see either a default profile icon or your profile picture (if you have uploaded one).
Profile Menu: Clicking the profile icon will open a menu that includes options like:
Theme: Change the visual theme of your interface.
My Profile: Access your profile settings.
Sign Out: Log out from the ChatNGO platform.
From the My Profile option, you can access several key areas, which we will cover in the following sections, including setting your name, uploading a profile picture, and managing your account details.
If you are part of multiple organizations or collaborative projects, ChatNGO makes it easy to switch between different workspaces without needing to log out and log back in. This allows you to seamlessly manage your work across different nonprofits or projects with the same login.
Top-Left Icon
The Images section in ChatNGO allows you to easily access and manage AI-generated images shared within your workspace. Whether you are reviewing images created for a project or just browsing through visual assets, this section provides a centralized location for all the images generated in your team chats.
To navigate to the Images section:
Click on
In the Profile Settings section, users can manage personal details such as their profile picture, and display name. Here is a step-by-step guide on updating these settings:
1. Access Profile Settings
Click on your profile icon (bottom left of the platform).
In the drop-down menu, click My Profile.
Folder permissions in ChatNGO provide flexibility to control who can access, view, or modify content within a folder. This ensures that the right team members have the appropriate level of access to collaborate efficiently while protecting sensitive information.
Sharing a Folder:
The Folders section in ChatNGO allows you to efficiently organize your chats ensuring your workspace remains clutter-free and easy to navigate. Folders help categorize your content, making it easier for teams to locate important information, collaborate on projects, and share relevant files with the right individuals.
Create Folders for Organization
Managing your folders effectively involves being able to rename, move, or delete them as needed. Here are the steps for each action:
To rename a folder, follow these steps:
Navigate to the Folder: Locate the folder you wish to rename within the "Team" or "Private" folder section in the "Chat" tab.
The Chat Navigation section provides an overview of how to manage and navigate your chats within ChatNGO. Whether you are communicating with your team or holding private discussions, this section will guide you through key features like switching between Team and Private chats, accessing recent conversations, starring important chats, and finding archived messages.
Team vs Private Chats: Learn the difference between Team Chats, which are visible to everyone in the workspace, and Private Chats, which are restricted to selected members.
The reporting feature in ChatNGO provides workspace admins and managers with insights into how the platform is being used. Reports can help you measure activity, user engagement, and the overall usage of your workspace. This functionality can be essential for understanding how effectively teams are collaborating and communicating, providing data-driven insights for improvement.
Monitor Team Activity: Keep track of message volume and the number of active users to ensure everyone is engaged and collaborating efficiently.
At any time, click on the ChatNGO icon located at the top left corner of your dashboard. This will take you to the workspace selection screen.
Workspace Selection
On the workspace selection screen, you will see a list of all the workspaces you are a part of. Simply click on the workspace name you want to access, and you will be taken to that workspace.
Workspace URL
Each workspace also has a unique URL that you can use to directly access it. For example: team.chat.ngo/your-workspace-name. Bookmarking these URLs can be a quick way to access your workspaces.
You may need to switch between workspaces if:
You are part of multiple organizations: For example, you volunteer for one nonprofit and are a board member for another. You will have separate workspaces for each organization.
Collaborative Projects: If you are involved in a project that brings together multiple nonprofits, you may have access to a dedicated workspace created specifically for that collaborative project.
Scroll down to find the gallery titled Generated Images.
In this gallery, you'll be able to view and manage all the AI-generated images shared within the workspace.
Shared Chats: Understand how to control access to Team Chats and share them with the right members based on your collaboration needs.
Recent Chats: Easily access your most recent conversations to stay updated on ongoing discussions.
Starred Chats: Keep track of important conversations by starring them for easy access later.
Archived Chats: Move older or less active chats to the archive to reduce clutter while keeping them available for future reference.
Images: Find and manage all the images generated or shared in your chats through the Generated Images gallery in Team Chats.
This will take you to the Profile Information page.
2. Update Profile Picture
To change your profile picture, click on the Upload Image icon.
Select an image from your device, and it will update your profile picture.
3. Update Name
You can modify your first and last name in the provided fields.
After making any changes, click Save to apply the updates.
5. Additional Settings
There's an option to enable or disable Folders Tree View Navigation, which helps in organizing the workspace by displaying it as a tree view for easier navigation.
Once you’ve made your changes, click Save to finalize your updates.
Go to the folder you wish to manage.
Click on the three dots (more options) next to the folder name.
Select Share from the dropdown menu.
Choosing Access Levels: When sharing a folder, you can set permissions for the whole workspace or individual users. The options include:
Full Access: Users can add, modify, or delete chats in the folder.
View Only: Users can view the content but cannot modify it.
Only Me: The folder is private, and only you have access.
Assigning Permissions to Specific Users:
You will see a list of workspace members.
For each user, you can assign one of the following permissions:
Inherit (Can Edit): Follows the general workspace permission settings for the user.
No Access: The user cannot see or interact with the folder.
Can View: The user can view the contents but not make changes.
Can Edit: The user can add or modify content within the folder.
Permissions for Entire Workspace: You can also set permissions for the entire workspace, allowing all users to have a unified access level. These options include:
Only Me: Limits access to yourself only.
View Only: Allows all workspace members to view the folder.
Full Access: Provides all workspace members with full control over the folder contents.
Users can create folders to store chats in a more structured way. For example, you can create separate folders for projects, teams, or specific departments within your organization.
Shared and Private Folders
Shared Folders: Accessible by multiple team members, allowing for collaborative work. These folders are visible to other members of the workspace, depending on permissions set by the folder creator.
Private Folders: Only visible to the user who created them. These folders are great for personal work or content that doesn’t require collaboration.
Folder Permissions
Permissions allow you to control who can access and manage folders. Users can set permission levels for each folder, ensuring that only authorized team members can view or edit content.
Subfolders for Deeper Organization
To further organize your content, you can create subfolders within your main folders. This allows for a hierarchical structure, making it easy to store and find content related to specific tasks, teams, or projects.
Organize Work by Projects: Create project-specific folders to keep all related chats in one place.
Team-Specific Collaboration: Shared folders can be created for specific teams, allowing team members to collaborate efficiently and share resources.
Track Ongoing Conversations: Use folders to organize chats that are part of long-term projects, making it easy to reference previous conversations and decisions.
Click on the Three Dots Next to the Folder: Hover over the folder name and click on the three dots (more options) next to it.
Select 'Rename': From the dropdown menu, select 'Rename.'
Enter the New Name: Type the new name for the folder and click Save to confirm the change.
You can move folders using two methods: drag-and-drop or by selecting the move option.
Drag-and-Drop Method
Locate the Folder: Find the folder you want to move within your folder list.
Click and Hold the Folder Name: Select the folder by clicking and holding on its name.
Drag the Folder to the New Location: Move it to the desired location within your folder structure, either to another folder or a subfolder.
Release to Drop: Once you have positioned the folder in the correct location, release it to complete the move.
Move Option Method
Locate the Folder: Find the folder you want to move within your folder list.
Click the Three Dots Next to the Folder: Open the more options menu by clicking the three dots next to the folder name.
Select 'Move': From the dropdown, click 'Move.'
Select the New Location: Choose the new location for the folder from the available list of folders and subfolders.
Click 'Move': Confirm the move by clicking Move.
Deleting a folder is a permanent action, so ensure the folder is no longer needed before removing it.
To delete a folder:
Go to the Folder: Navigate to the folder you wish to delete.
Open the Options Menu: Click on the three dots next to the folder name to bring up the options menu.
Select 'Delete': Choose 'Delete' from the dropdown menu.
Confirm the Deletion: A prompt will ask you to confirm the deletion. Click Delete to permanently remove the folder and its contents.
Assess Workspace Usage: Gauge the effectiveness of your workspace by tracking the number of chats and messages.
Data-Driven Insights: Gain valuable insights into how different teams or users are interacting, allowing you to optimize workflows and collaboration efforts.
Stay tuned for additional reporting features that will expand the depth of analysis available in future updates.
General
Configure basic settings for your workspace, such as the name, workspace URL, and workspace icon. These elements help identify and customize your workspace.
Members
Manage who has access to the workspace, assign roles (Admin, Manager, User), and invite new members. This section ensures that the right people have the appropriate permissions within the workspace.
Models & Credentials
Set up and manage the AI models available to your team. You can add API keys for different AI models (such as OpenAI, Anthropic, Perplexity), and enable or disable models based on your team’s needs.
Custom Instructions
Create and manage custom instructions that can be applied to AI prompts. These instructions help fine-tune the behaviour of AI models, allowing them to respond in line with your organization's specific tone, style, or guidelines.
Click on your workspace name in the top navigation bar.
Select Reports from the dropdown menu.
From the Side Menu:
Click the More icon in the left-hand menu (represented by three dots).
Select Reports from the options.
Once you have accessed the Reports page, you will see the following metrics:
Total Number of Messages: This metric shows the total count of all messages sent within your workspace. It gives you an overview of the communication activity in your workspace.
Total Number of Chats: This section displays the total number of active chats, providing insight into how much collaboration is happening across your teams and projects.
Active Users: A bar graph illustrates the number of unique active users over time. This is a useful tool for tracking engagement and seeing when users are most active.
User Engagement: In this table, you can see each user listed by name, along with their number of chats and messages. This helps assess the level of contribution from each member of your team.
This basic reporting setup is designed to give you a quick glance at user activity and engagement in your workspace, with more detailed and advanced reports coming soon.
Admins and managers can invite new members to the workspace by entering their email addresses. Once invited, users will receive an email prompting them to join the workspace.
You can also create an invite link that can be shared with prospective members to join the workspace directly.
Assign or Change Roles
Every member in the workspace is assigned a role: Admin, Manager, or Member. These roles determine the level of access and permissions the user has within the workspace.
To assign or change a role, select the user from the list, click on their current role, and choose the appropriate role from the dropdown menu.
Remove Members
If a member no longer needs access to the workspace, admins can remove them. Simply click on the three dots and select 'Delete Member', and they will no longer have access to the workspace.
From your workspace, click on Workspace Name on the top menu.
Select Workplace Settings from the dropdown menu.
Select Members from the left-side menu.
In the Invite by Email section, enter the email address of the person you would like to invite and assign them a role before clicking Send Invite.
Alternatively, you can use the Invite Link option to create a link that can be shared with others.
The roles assigned to members determine what they can do within the workspace. Visit the Roles and Permissions section for a detailed explanation of what each role can do.
ChatNGO provides two main types of chats to facilitate collaboration and personal work: Team Chats and Private Chats.
Shared Visibility: When a Team Chat is created, it is automatically shared with the entire workspace by default. This means all members of the workspace can view and contribute to the conversation.
Collaboration: Team chats are designed for open communication within an organization or team. They are ideal for collaborative projects, brainstorming sessions, or when several members need to align on a task.
Adjustable Sharing: While Team Chats are shared with the full workspace by default, you can adjust sharing settings to limit access to specific users. We will cover this in the next section.
Transparency: Team chats ensure that everyone is on the same page, as conversations are open and accessible to all team members unless adjusted otherwise.
Personal Use: Private chats are for individual use, and only the user who started the chat can access the conversation. These chats are perfect for tasks like drafting personal notes, researching ideas, or conducting individual tasks that do not require input from others.
Privacy and Confidentiality: Since private chats are restricted to the user, they are ideal for working on sensitive information or brainstorming ideas before presenting them to a wider group.
Team to Private: You can move a Team Chat to a Private Chat if you no longer wish to share the conversation with the entire workspace. This can be useful for more confidential or individual work that does not require team input.
Private to Team: Likewise, you can turn a Private Chat into a Team Chat at any time, making it visible and accessible to the full workspace or selected members. This allows flexibility in how you manage your conversations.
Use Team Chats when working on projects that require collaboration, feedback, or shared input from your colleagues or team.
Use Private Chats for solo tasks or when you are handling information that does not need to be shared with others in your workspace.
To integrate OpenAI or Anthropic models into your workspace, you will need an API key for each service. Below are the steps to generate API keys from both OpenAI and Anthropic, along with important billing and usage considerations.
Step 1. Account Setup
Visit and create an account or sign in with existing credentials.
If you're a new user, verify your account through the confirmation email that OpenAI sends.
Step 2. Generating the API Key
Once logged in, click on your profile icon in the top-right corner of the page.
Select "Dashboard" from the dropdown menu.
In the left-hand menu, click "API Keys."
Click "Create new secret key."
Step 3. Important Setup Steps
Billing: Navigate to the Billing section to configure your billing preferences and set usage limits to manage monthly expenses.
Usage Limits: Ensure you set a cap on your API usage to avoid unexpected costs.
Terms of Service: Review OpenAI’s usage policies and terms of service to understand compliance requirements.
Step 1. Account Setup
Visit and either create a new account or sign in using Google authentication for a quick login.
Step 2. Generating the API Key
Once logged in to the dashboard, click on "Settings" on the top menu, and then "API Keys" on the left menu.
Click the orange "Create Key" button.
Provide a name for the key (e.g., the project name it will be used for).
Important: Copy and securely store the key immediately, as it won’t be visible again after you close the window.
Step 3. Usage and Billing
Billing Configuration:
New accounts may receive initial credits for testing purposes.
You'll need to provide a phone number to claim these credits.
Without credits, you will need to select a payment plan:
Ensure your API keys are stored securely, as you won’t be able to view them again after initial generation.
Review both OpenAI and Anthropic’s usage policies, rate limits, and billing options to manage costs effectively.
Be aware of rate limits and configure usage to avoid exceeding limits, which can disrupt your service.
To start a new chat in ChatNGO, users have two options to initiate the process:
Click the Green Plus Button:
Located on the left sidebar of the ChatNGO interface, clicking this green "+" button will instantly open a new chat thread.
Click the Plus Icon in the Top Right of the Left Sidebar:
Another option is to navigate to the top right corner of the left section where a "+" icon is visible. Clicking this icon also starts a new chat.
Once the new chat is opened, the Chat Bar becomes your primary tool for interacting with the AI models. Here, you can:
Enter your text or questions for the AI to respond to.
Attach images or files to enhance the context of your request.
Select from available prompts using the .
Use for more tailored interactions with the AI.
The system will automatically save your chat thread under chats depending on your current selection, allowing for further collaboration with team members or private workspaces.
To add an API key to your workspace and enable AI usage, follow these steps:
Only users with the Admin role can add or manage API keys in the workspace. If you do not have admin privileges, contact your workspace admin.
Click on the Workspace Settings option from the left menu.
Navigate to the Models & Credentials section.
Click on the + Add model button.
From the dropdown menu, select the API provider you want to add, either OpenAI or Anthropic.
In the API Key field, paste your API key.
Click Save.
Once the API key is added successfully, the available models for that provider, such as GPT-4o or Claude 3.5, will be displayed.
You can toggle each model on or off depending on the needs of your workspace.
In the Default Models section, you can select a default model for various functions like Chat, Vision, or Image Generation.
Choose the desired default model for each function from the dropdown.
Click Save to confirm your selection.
In the Theme section, users can customize the appearance of their ChatNGO interface by switching between different themes or following the system settings for appearance. Here is how you can modify these settings:
1. Access Theme Settings
Click on your profile icon located in the bottom left of the platform.
From the drop-down menu, click Theme.
2. Choose Your Theme
You have three theme options to choose from:
System: The platform will automatically follow your device’s default light or dark theme settings.
Light: The interface will be displayed in light mode, providing a brighter appearance.
Dark: The interface will be displayed in dark mode, reducing eye strain and providing a more muted, darker look.
3. Selecting a Theme
Simply click on the desired theme option (System, Light, or Dark) to activate it immediately.
Once you select the preferred theme, your interface will automatically update to reflect the changes. There is no need to save manually—the theme changes are instant.
Starred Chats are conversations that you have marked as important. These chats are stored in a separate section, allowing you to focus on them without sorting through all your active chats. Any chat, whether it is a Team Chat or a Private Chat, can be starred.
Easy Access: Starred chats appear in their own dedicated section, making it simple to jump straight to important conversations.
Organization: This feature helps you organise your workspace by keeping vital discussions separated from general or less urgent chats.
Flexible: You can star and unstar chats as needed, giving you complete control over which conversations you want to focus on.
To star a chat:
Click on the three dots (•••) next to the chat name in the list.
Select Add to Starred from the dropdown menu.
Once a chat is starred, it will automatically appear in the Starred Chats section for quick access.
To remove a chat from your Starred Chats:
Go to the Starred Chats section.
Click on the three dots (•••) next to the starred chat you wish to remove.
Select Remove from Starred.
The chat will no longer appear in your starred list but will remain in its original location.
Prioritize Important Conversations: Use Starred Chats to keep key conversations at the top of your list, so you never miss an update on the most critical discussions.
Update Your Starred Chats Regularly: As priorities shift, remember to update your Starred Chats to reflect your current focus areas.
Limit Overuse: Avoid starring too many chats, as this can clutter your Starred Chats list and reduce its effectiveness as a tool for prioritization.
At this time, ChatNGO only supports attaching images to your chat threads. Other file types such as documents or files from cloud storage (Google Drive, Microsoft OneDrive) are coming soon. Here is how you can attach images:
Locate the Attach Button:
On the chat bar, next to the text input field, click the "Attach" button.
Select Image:
From the menu, select the option labeled Image. This will allow you to upload an image from your local device.
Choose the File:
Once selected, your computer’s file explorer will open. Choose the image file (such as PNG or JPEG) you want to upload.
Send Image:
After selecting the file, it will be uploaded into the chat. You can add context to the image by typing a message before clicking the send button (the green arrow) to complete the upload.
While only image uploads are currently available, future updates will introduce the ability to upload other file types (such as documents) and direct integrations with Google Drive and Microsoft OneDrive.
Archived Chats are conversations that have been removed from your active chat list but are still saved within the workspace. You can access archived chats at any time, but they will not appear in your main list of active conversations unless they are unarchived.
Declutter Your Workspace: Archiving helps keep your active chat list clean and focused on current discussions.
Easy Retrieval: Even though a chat is archived, it can be accessed whenever needed, ensuring that no important information is lost.
Reversibility: Archived chats can be unarchived and moved back to your active chat list at any time.
To archive a chat:
Click on the three dots (•••) next to the chat name in the chat list.
Select Archive from the dropdown menu.
Once a chat is archived, it will no longer appear in your active chat list but will be stored in the Archived Chats section.
To bring a chat back into your active list:
Go to the Archived Chats section by clicking on More on the left menu and then on Chat Archive .
Click on the three dots (•••) next to the chat you want to unarchive.
Select Restore.
The chat will reappear in your active chat list, allowing you to continue the conversation as needed.
Archive When Necessary: Archive chats that are no longer active but may still hold valuable information. This keeps your active workspace clean and efficient.
Unarchive for Ongoing Projects: If a project or discussion becomes relevant again, simply unarchive the chat to pick up where you left off.
Do Not Over-Archive: Make sure not to archive chats you might frequently need. Archived chats require an extra step to access, so use this feature for conversations that are genuinely inactive.
The Custom Instructions section allows Admins to create or manage pre-defined instructions that can be enabled on chats by users across the workspace. Custom instructions allow you to preset guidelines for how the AI should respond in different scenarios. This can include persona details, formatting requirements, language settings, tone, and more.
To create a new custom instruction:
Go to Workspace Settings > Custom Instructions.
Click on Create New Instruction.
Enter a Name for the instruction (e.g., “Professional Tone”).
Select a Type from the dropdown (e.g., Persona, Brand Voice, Language, Writing Style).
Add the Custom Instructions in the text box (e.g., “Always use a professional and formal tone”).
Save the instruction to make it available in your workspace.
Once custom instructions are created, they can be applied to any chat or prompt to ensure the AI adheres to the predefined guidelines. Users can choose from existing instructions or create new ones as needed.
Instructions are categorized into different types to help organize and apply them more efficiently:
Persona: Define a specific persona for the AI, such as “Behavioral Scientist” or “Project Manager.”
Brand Voice: Set the tone and style that aligns with your organization’s communication guidelines.
Language: Define the language or dialect the AI should use (e.g., US English, British English).
Writing Style: Specify how the AI should structure its writing, such as “Concise and Clear” or “Engaging Narrative.”
Admins can view, edit, or delete existing instructions from the custom instructions dashboard. This ensures that the workspace stays up to date with the latest communication or project guidelines. To edit or delete an instruction, click on the three dots next to the instruction and choose the appropriate action.
Team Personas: Define different personas based on your organization’s departments (e.g., marketing, fundraising) to ensure the AI tailors responses appropriately for each team.
Brand Consistency: Set custom instructions for brand voice to maintain consistency across all AI-generated content.
Project-Specific Guidelines: Apply unique instructions for special projects that may require a particular tone, formatting, or response structure.
To generate an image in ChatNGO, follow these steps:
Click the "Generate Image" Icon: On the top menu bar within your chat workspace, click the icon that looks like a picture with a plus sign.
Enter Image Description: Once the image generation panel opens, you will be prompted to enter a description of the image you would like to generate in the text box provided.
Choose Image Orientation: Select your preferred image orientation from the options available:
Square
Portrait
Landscape
HD Option: You can toggle the HD option on or off, depending on whether you would like the image in high-definition.
Generate: Once you have entered your description and selected your settings, click the arrow icon to generate the image. The system will create the image based on your input using the DALL-E 3 model.
Viewing Generated Images: The images you generate will appear under the "Generated Images" in the left-hand panel. You can see a preview of each image there.
Once the image is generated, you can interact with it by clicking on the image for further actions such as regenerating, downloading or deleting.
At ChatNGO, we are committed to ensuring you have the resources and assistance you need to make the most out of the platform. Our support options are designed to help you troubleshoot issues, answer questions, and guide you through various features of the workspace
Available Support Channels
Office Hours:
Join our weekly office hours where you can ask questions, receive guidance, and get insights into the best ways to utilize ChatNGO for your organization. Check the Office Hours page for the schedule and details on how to join.
AI Search:
Quickly find answers to your questions using our AI-powered search tool. Simply enter your query in the top right 'Ask or Search' bar, and the system will provide instant suggestions and articles from our knowledge base and documentation.
Contact Information:
If you need direct support from our team, you can always reach out via our page. Here, you wiill find details on how to get in touch with us for more in-depth support.
API keys are essential tools for integrating AI models, such as OpenAI and Anthropic, into your workspace. They serve as a unique identifier that authorizes your workspace to access AI services, enabling your team to leverage advanced AI capabilities within ChatNGO. Here is a breakdown of what API keys are, how they centralize AI usage and billing, and why they are a cost-effective way to provide AI access across your team.
The Chat section in ChatNGO is the heart of the platform, allowing users to interact with AI models to generate insights, complete tasks, and collaborate with others in the workspace. Whether you are drafting a report, analyzing data, or brainstorming ideas, the chat interface provides a dynamic space where you can leverage the power of AI in real-time.
The Prompt Library is a central hub where users can access and organize various prompts for ChatNGO. Prompts allow you to interact with the AI in structured ways, and the library provides a convenient space to save, edit, and reuse prompts tailored for your organization.
To access the Prompt Library, click on the “Prompts” button at the bottom right of the chat window. You will then see three sections:
In the Change Password section, users can update their password to ensure the security of their ChatNGO account. Here is how to change your password:
1. Access Your Profile Settings
Click on your profile icon at the bottom left of the platform.
In the drop-down menu, click on My profile.
2. Navigate to Password Settings
Shared Chats in ChatNGO allow you to collaborate with selected members of your workspace in a controlled and organized manner. These chats help streamline communication and ensure that everyone involved in a project or discussion has access to the necessary information.
Shared by Default in Team Chats: When a Team Chat is created, it is automatically shared with the full workspace by default. However, you can adjust who has access to the chat depending on the specific needs of your project or conversation.
The Recent Chats feature in ChatNGO provides an easy way to keep track of your most current conversations. Whether you are juggling multiple projects or collaborating with different teams, this section allows you to quickly return to your ongoing discussions.
The Recent Chats tab displays the most recently active conversations. This includes both Team Chats and Private Chats, helping you resume important conversations without needing to search through the entire workspace.
The Models & Credentials section allows Admins to configure the AI models available to the workspace. This is where you can add API keys for different AI providers, enable or disable models, and adjust the model-specific settings such as temperature and context window. Having control over AI models ensures that your team can manage costs and leverage the right tools for their projects and tasks.
Add API Keys
An API key is a unique string of characters that allows your workspace to securely communicate with external AI services, such as OpenAI or Anthropic. Instead of each team member needing a separate API key, a single key is used for the entire workspace. This centralized approach simplifies AI management and billing by:
Allowing all members of the workspace to use AI through a shared key.
Tracking usage collectively, not individually, so that the workspace pays for AI usage as a whole rather than on a per-member basis.
Each provider requires a unique API key, meaning you will need one API key for OpenAI and another for Anthropic. A single API key from a provider grants access to all the models that provider offers.
1. Cost-Effective AI Access
Single Payment Structure: Using one API key for the entire workspace means that you only pay for the actual AI usage rather than on a per-user basis. This is particularly cost-effective for larger teams.
No Per-Member Charges: The whole team can access AI without needing to set up individual accounts or keys, reducing administrative work and potential fees.
2. Centralized Usage Tracking
Monitor AI Usage: Since all usage is routed through the workspace’s API key, you can easily track the total amount of AI being used. This helps in budgeting and adjusting based on actual needs.
Set Usage Limits: You can set limits to avoid exceeding your budget, ensuring that sudden increases in usage do not lead to unexpected costs.
When using AI services, your workspace's access to models is managed by usage tiers, each with specific rate limits. These tiers determine the amount of AI processing you can access, such as the number of requests or tokens allowed per minute or day.
Rate Limits: Each tier has a cap on the number of requests per minute, tokens per minute, or tokens per day. Exceeding these limits may result in request throttling or "429 Error" messages indicating too many requests.
Tiers Vary by Provider: OpenAI and Anthropic offer multiple usage tiers, with higher tiers providing more generous limits. Access to these higher tiers may require a minimum period of usage or reaching specific spending thresholds.
Usage Limits: Usage limits are determined by the AI model providers (such as OpenAI and Anthropic) and are not controlled or set by ChatNGO.
Tier: The selected tier affects only the rate limits (such as requests per minute) and does not impact the cost per token used.
1. Monitor Billing Regularly
Track Usage: Use the dashboards provided by OpenAI and Anthropic to keep an eye on how much AI is being used in your workspace.
Set Limits: To prevent unexpected costs, consider setting usage limits on API calls.
2. Check Model Costs
Model Cost Variations: Different models (e.g., GPT-4o, Claude 3.5 Sonnet) have different costs depending on complexity and data processed. Simpler tasks can often be handled by more cost-efficient models, while complex analysis may require more advanced, higher-cost models.
Enable Models Strategically: Avoid enabling all models if some are cost-prohibitive or unnecessary. Admins can educate users on which model is best for specific tasks.
3. Configure Billing and Payment Plans
Both OpenAI and Anthropic offer flexible pay-per-usage billing options. Be sure to configure these settings based on your workspace’s needs to optimize your budget.
ChatNGO supports dynamic, interactive conversations with AI models like GPT-4o and others, which can help users with a wide range of tasks, from writing and summarizing to generating ideas and analyzing information.
Customizable Language Models
Users can choose from a variety of AI models, each with different capabilities. Depending on the workspace settings, you can select the model that best suits your needs. These models are configured by workspace admins but are easily accessible to users through the chat interface.
Threaded Conversations
Chat threads allow you to organize your conversations based on topics or tasks. This makes it easy to revisit previous chats for reference or to continue where you left off. You can have multiple threads going simultaneously, with each conversation saved and categorized within your workspace.
Shared and Private Chats
Shared Chats: These conversations can be accessed by other team members in your workspace, allowing for collaborative work. This is useful for teams that need to follow the same conversation, provide feedback, or continue a task started by another member.
Private Chats: These chats are only visible to the user and are perfect for working on personal tasks or sensitive information.
Attachment and File Sharing
In the chat interface, users can upload and attach images to assist the AI in understanding the context of a conversation. This feature allows you to leverage AI’s vision capabilities for image-based queries.
Prompts and Custom Instructions
You can enhance the AI’s responses by using predefined prompts or custom instructions. These features enable you to structure your chats in a way that aligns with your project’s needs, ensuring the output is relevant and actionable.
Real-Time Collaboration
For teams working together, ChatNGO offers real-time chat and collaboration features, where multiple users can contribute to a conversation simultaneously. This makes it easy for teams to brainstorm, plan, and execute projects effectively.
Brainstorming Ideas: Collaborate with AI to generate creative ideas, marketing strategies, or campaign concepts.
Drafting Documents: Have AI assist with writing emails, reports, or presentations by providing content or editing suggestions.
Data Analysis: Input data and have the AI summarize, interpret, or provide insights based on the information given.
Team Collaboration: Share a chat thread with your team to gather feedback or allow others to pick up where you left off.
Get more ideas by checking out the prompt library.
Team: Shared prompts that are accessible to all members of your workspace.
Private: Prompts that are only visible to you.
System: Predefined prompts provided by ChatNGO. These prompts are available to all users and updated periodically based on feedback and requests from the community.
Once inside the library, you can browse folders and prompts that have been organized into categories. For example, under the System tab, prompts are grouped into categories like Social Media, Fundraising, or Meeting Management.
Selecting a Prompt: Simply click on a prompt to use it. If the prompt requires input from you, such as a name or specific piece of information, a form will appear where you can fill in the required fields.
Filling in Variables:
Prompts may contain variables (fields marked by curly brackets, such as {{Donor Name}}). These variables allow the prompt to be dynamic and tailored to different scenarios. When a prompt is selected, you will be prompted to fill in these variables before proceeding.
Tweaking the Prompt: Once the prompt is filled out, it will be inserted into the chat window. You can make any final tweaks to the text before hitting “Send” to interact with the AI.
Exploring System Prompts: System prompts provided by ChatNGO cover a range of tasks, such as generating social media posts, summarising meetings, and drafting donor emails. These are regularly updated and can be a great starting point for new users.
Creating a new prompt allows users to build a customized piece of reusable text that can include dynamic variables. These variables are placeholders that will prompt the user to input values each time the prompt is used, ensuring flexibility across different contexts.
Here is how you can create a new prompt:
Click on the “+ New Prompt” Button: Whether you are in the Team or Private section of the Prompt Library, click on the + New Prompt button to open the prompt creation window.
Enter the Prompt Name: In the Name field at the top of the window, enter a descriptive name for your prompt. This will make it easier to locate the prompt in the library when needed.
Add a Description (Optional): In the Description field, you can add a short explanation or context for the prompt. This is helpful if others in the team will use it, or if you need a reminder of the specific use case for the prompt.
Add Instructions:
The Instructions field is where you will enter the body of the prompt itself. You can use Markdown syntax here for formatting, such as bolding or adding bullet points. If you need to add dynamic fields (variables), use the {{variable_name}} syntax. For example, {{Donor Name}} will prompt the user to enter a specific name when the prompt is used.
Preview the Prompt: As you add your instructions, a Preview will automatically update on the right side of the window, allowing you to see how the final prompt will look.
Save or Save and Use: Once you have completed the prompt, you have two options:
Save: This will save the prompt to your library for future use.
Save and Use: This will save the prompt and immediately insert it into the active chat window, allowing you to start using it right away.
The prompt library also allows you to organize prompts into folders. For example, if you have several prompts related to Fundraising, you can create a Fundraising folder and place all relevant prompts inside for easy access.
Creating Folders: Click on the Create Folder button to organize prompts. You can have different folders for different types of prompts, such as Media Releases, Donor Emails, or Team Meetings.
Folder Access: Folders can be shared across the team in the Team tab, or kept private for your own use in the Private tab. System folders are available to all users and cannot be edited.
Click on the New Folder Icon: In the top-right corner of the folders view, click on the "New Folder" icon.
Enter a Name for the Folder: Type in the folder name.
Click 'Create': Finalize the folder creation by clicking Create.
For deeper organization, you can create subfolders within existing folders. Subfolders allow you to break down your content into smaller, more manageable categories.
To add a subfolder:
Navigate to the Parent Folder: Find the folder where you want to create a subfolder.
Click the New Folder Icon: Select the New Folder Icon next to the three dots.
Enter a Name for the Folder: Type in the subfolder name.
Create the Subfolder: Finalize by clicking Create.
Keeping your workspace organized often involves moving chats and files between folders. You can do this using two methods: the drag-and-drop method or the more options menu.
Drag-and-Drop Method
Click and Hold the Chat: Click and hold on the chat thread.
Drag to the Target Folder: Drag the chat to the appropriate folder in the folder list.
Release to Move: Drop the item into the folder by releasing the mouse.
More Options Method
Click the Three Dots Next to the Item: For the chat or file you want to move, click the three dots next to the name.
Select 'Move to Folder': From the menu, choose Move to Folder.
Choose the Target Folder: Select the destination folder from the list and click Move.
Assign color codes to your folders for quick visual identification. This is especially useful for distinguishing between different projects or departments.
To add a color code:
Click the Three Dots Next to the Folder: Access the folder options.
Select 'Color': Choose from the available color options to assign a color to the folder.
Save: The folder will now display the selected color.
Use Clear Naming Conventions: Make sure your folder names reflect the content inside, such as "Project A – Reports" or "Fundraising – Meeting Notes."
Organize by Function or Department: Group folders by function, project, or department to make navigation easier for users.
Regularly Review and Clean Up: Archive or delete folders that are no longer in use to keep the workspace uncluttered.
Use Subfolders Sparingly: While subfolders can be helpful, avoid overcomplicating the structure to keep everything intuitive and accessible.
Once in the Profile section, you will see the Password option in the left-hand menu. Click on it to proceed.
3. Change Your Password
In the Password settings:
Enter your current password in the Current password field.
Enter your new password in the New password field.
Confirm the new password by re-entering it in the Confirm new password field.
4. Save Changes
Click the Save button to update your password.
Your password will now be updated. Be sure to remember your new password for future logins. If you forget your password, you can use the "Forgot Password" option on the login screen to reset it.
Quick Access: All your latest conversations are grouped here for easy access. Simply click on any chat from the list to jump right back into the discussion.
Organized by Activity: Chats are sorted based on the most recent activity, ensuring that your most important or active conversations are always at the top.
Team & Private Chats: The Recent Chats section will display both Team and Private chats. This way, you can quickly see where the latest messages are coming from, regardless of the type of chat.
Archiving: If a conversation is no longer active but you want to keep it for reference, you can move it to the Archived Chats section, which keeps it out of your Recent Chats list but allows for future reference.
Stay Organized: Use the Recent Chats list to manage your active conversations, but archive those that no longer need immediate attention.
Prioritize Communication: Keep an eye on this section for real-time updates in the conversations most important to your work, ensuring nothing falls through the cracks.
The name of the workspace helps identify it among other workspaces you may have access to. This is usually the name of the organization or the project the workspace represents.
To change the workspace name, simply enter the new name in the designated field and click Save.
Workspace Icon
Customize your workspace by uploading a unique icon or logo. This helps team members quickly identify the workspace, especially if they are part of multiple workspaces.
Click on Upload Image and select the desired icon from your device.
Workspace URL Slug
The workspace URL slug is a unique identifier for your workspace. It forms part of the URL that team members will use to access the workspace.
For example, if your workspace is called “ABC NonProfit,” the URL might look like team.chat.ngo/abc-nonprofit. You can edit this slug in the General Settings and click Save to confirm changes.
Public Chats
Enable or disable public chats in the workspace. Public chats allow users to share chats with people outside of the workspace.
Personal Chats
Manage the ability for team members to create private chats that are only visible to them.
Admins can add API keys for supported AI models such as OpenAI, Anthropic, and Perplexity. Adding these credentials enables your workspace to access the models for use in chats and prompts.
To add an API key:
Select the provider (OpenAI, Anthropic, Perplexity) from the dropdown menu.
Enter the API key in the provided field.
Click Save to activate the model for your workspace.
Enable or Disable Models
Once API keys are added, admins can choose which models are available for workspace members. For example, you may enable GPT-4o for complex tasks but disable GPT-3.5 if it is not needed.
To enable or disable a model:
Toggle the switch next to the model name.
Enabled models will be available in the chat interface and prompts section for all workspace members.
Adjust Model Settings
Admins can fine-tune specific settings for each enabled model, such as temperature and context window:
Temperature: This controls how creative or focused the AI model is. A higher temperature (e.g., 1.0) makes the model more creative, while a lower temperature (e.g., 0.2) makes it more deterministic.
Context Window: This defines how much text the model can keep in context during conversations or tasks. Larger context windows allow the model to remember more information from previous interactions but increases the cost.
OpenAI: Offers models like GPT-4o and GPT-o1-preview, which are widely used for generating text, performing complex tasks, and answering questions.
Anthropic: Provides Claude, a family of models that focus on safe and helpful AI outputs.
Perplexity: Known for models that are particularly good at research and knowledge retrieval tasks.
Go to Workspace Settings > Models & Credentials.
Add or update your API keys for the desired providers.
Toggle the models you want to enable or disable.
Adjust model settings as necessary to suit your workspace needs.
Important: Copy and save the key immediately, as it will not be visible again after you close the window.
Build: The standard self-service plan.
Scale: An enterprise plan requiring sales team contact.
Rate Limits: The API is subject to rate limits, measured by requests per minute, tokens per minute, and tokens per day. Exceeding these limits will result in a 429 error.
Pricing: Anthropic operates on a prepaid credit system. Costs vary based on the complexity of each API call.
Controlled Access: Although Team Chats are shared workspace-wide, Shared Chats allow you to modify access more selectively. You can invite specific members and remove others as necessary, giving you control over who participates.
Flexibility: You have full control over which workspace members can view and participate in a Shared Chat. If certain discussions are only relevant to a subset of people, you can adjust permissions to reflect that.
Convert to Private: If a Shared Chat no longer needs to be accessible to multiple members, you can convert it to a Private Chat. This restricts access to yourself or select members.
Convert Private to Shared: If a Private Chat evolves into a discussion that requires input from more people, you can convert it into a Shared Chat and invite others to join.
Find the Chat Thread: Navigate to the chat thread you want to share.
Open Sharing Options: Click the three dots (•••) next to the chat thread's title and select Share from the dropdown menu.
Set Public Access: In the sharing settings, you will see options for adjusting the public access of the chat:
No Access: This means only workspace members with proper permissions can view or interact with the chat.
Public: Selecting this option will allow anyone with the link to view the chat thread, even if they do not have a ChatNGO account. This functions similarly to how a Google Doc can be publicly shared.
Only Me: This restricts the chat to private, meaning only you can access and modify it.
Set Permissions for Workspace Members: Below the public access settings, you will see a list of workspace members and their current access permissions. You can adjust each member's permissions individually:
No Access: The user cannot see or interact with the chat.
Can View: The user can see the chat but not edit or add messages.
Can Edit
Project-Specific Discussions: Shared Chats are ideal for project discussions where only select members need access and updates.
Team Collaboration: For smaller teams working on specific tasks within a larger workspace, Shared Chats offer focused communication without cluttering the workspace for others.
Inter-Organization Collaboration: If multiple organizations or teams are working on a collaborative project, Shared Chats allow for targeted and effective communication.
Keep it Relevant: Only involve the necessary people in Shared Chats to avoid unnecessary confusion or information overload.
Review Permissions Regularly: As projects progress or change, regularly review who has access to your Shared Chats to ensure the right people are included or removed.
Selecting the right AI model for your tasks is crucial to ensure both efficiency and quality. Each model available in ChatNGO has unique strengths, suited for different types of tasks. Below is a guide to help you choose the best model for your specific needs, balancing between complexity, speed, and cost.
Description: GPT-4o is OpenAI's latest large language model, designed to handle multimodal inputs and outputs. It excels in complex reasoning and can process both text and visual inputs seamlessly.
When to Use:
Complex tasks that require advanced reasoning and multimodal understanding.
Strengths:
Advanced visual analysis capabilities.
Improved language understanding across 50+ languages.
Larger context window of 128K tokens for handling large tasks.
Limitations:
Currently supports text and image inputs with text outputs only.
May be overkill for simpler tasks due to its advanced capabilities.
Description: GPT-4o mini is a smaller, more efficient version of GPT-4o, designed for quicker responses at a lower cost while still maintaining high accuracy.
When to Use:
Applications requiring a balance between capability and efficiency.
Tasks needing quick responses and lower computational costs.
Strengths:
More cost-effective, being 60% cheaper than GPT-3.5 Turbo.
Supports multimodal inputs and outputs (text and images).
Outperforms GPT-3.5 in accuracy while maintaining speed.
Limitations:
Less powerful than full GPT-4o for very complex tasks.
Limited to text and image inputs with text outputs.
Description: Claude 3.5 Sonnet is a high-performance model designed by Anthropic, offering improved speed and accuracy, particularly in coding and reasoning tasks.
When to Use:
Tasks requiring graduate-level reasoning and undergraduate-level knowledge.
Coding tasks and multi-step workflows.
Visual reasoning tasks that require a fast yet thorough response.
Strengths:
Operates twice as fast as Claude 3 Opus.
Excels in visual reasoning and coding proficiency.
Offers improved understanding of nuance and complex instructions.
Limitations:
May not match the depth of analysis provided by Claude 3 Opus for very complex tasks.
Description: Claude 3 Opus is Anthropic’s most capable model, focusing on deep analytical capabilities for complex tasks.
When to Use:
Tasks that require detailed research and high-level content creation.
Projects demanding nuanced responses and in-depth analysis.
Strengths:
Superior analytical capabilities, excellent for generating detailed, high-quality content.
Ideal for handling complex and nuanced tasks.
Limitations:
Slower processing speed compared to Claude 3.5 Sonnet.
May be overkill for simpler, more straightforward tasks.
Description: Claude Haiku is the fastest model in the Claude 3 family, built for near-instantaneous response times.
When to Use:
Tasks requiring immediate responses.
Real-time interactions, quick analysis, or tasks such as content moderation and customer service inquiries.
Strengths:
Near-instantaneous response times.
Ideal for scenarios where speed is crucial, such as real-time interactions.
Effective for quick, straightforward tasks that do not require much depth.
Limitations:
Less depth and complexity in responses compared to Opus and Sonnet.
Not suitable for tasks requiring in-depth analysis or complex reasoning.
In summary, here’s a quick guide to choosing the right model:
GPT-4o: Best for highly complex, multimodal tasks needing deep analysis and reasoning.
GPT-4o mini: Ideal for efficient tasks requiring good capability at a lower cost.
Claude 3.5 Sonnet: Excellent for writing, reasoning, and coding tasks with quick responses.
Claude 3 Opus: Perfect for detailed research, high-level content creation, and in-depth analysis.
Claude Haiku: The fastest model for real-time responses, ideal for straightforward, quick tasks.
Use this guide to select the most appropriate model based on the complexity of your task, your speed requirements, and your budget.
Roles and permissions in ChatNGO determine the level of access and control users have within the workspace. There are three primary roles: Admin, Manager, and Member, each with different permissions. Understanding these roles is important for ensuring the right people have the appropriate level of access in the workspace.
Admin
Admins have full control over the workspace settings and members. They can:
Change workspace settings such as the name, URL slug, and icon.
Manage members (invite, remove, change roles).
Configure AI models by adding API keys, enabling/disabling models, and adjusting their settings.
Manager
Managers have limited administrative control and focus specifically on managing members. They can:
Invite new members, remove members, and change roles for existing members.
View workspace settings, but cannot make changes to them.
Member
Members have access to use the workspace and collaborate, but they cannot change any settings. They can:
Participate in chats, collaborate on projects, and use AI models provided by the workspace.
View shared resources and folders within the workspace.
Admins can change the roles of any member within the workspace:
Go to Workspace Settings > Members.
Find the user whose role you would like to change and click on their current role.
Select the new role from the dropdown menu (Admin, Manager, or Member).
The changes will be applied immediately.
Manage folders and permissions.
Create and manage custom instructions.
Admins have the highest level of access in the workspace and can make changes across all settings.
Managers cannot change AI model settings or other workspace configurations outside of user management.
Members cannot manage users or make any changes to workspace settings, AI models, or permissions.
When starting a new chat, choosing the appropriate AI language model ensures you receive the best responses based on your needs. ChatNGO offers several models, each with its own strengths.
Click on the Model Dropdown:
Below the chat bar, you will see a dropdown that shows the currently selected model (e.g., GPT-4o, GPT-4o Mini, Claude 3.5 Sonnet).
Click on this dropdown to view the available models.
Review Model Options:
The models available for selection include:
GPT-4o: Ideal for handling detailed and complex tasks with greater depth and reasoning.
GPT-4o Mini: A lighter, faster version of GPT-4o, suitable for simpler tasks or when quicker responses are needed.
Choose Your Model:
Select the model that best fits your needs by clicking on it from the dropdown list.
The selected model will be active for the current chat session and all future interactions within that thread.
Complex Tasks: Use GPT-4o for tasks that require deep analysis or brainstorming.
Faster Interactions: Use GPT-4o Mini when you need faster, more concise responses.
Creative Conversations: Select Claude 3.5 Sonnet for engaging in natural, creative discussions or for more conversational tasks.
Claude 3.5 Sonnet: Best for natural language understanding and creative tasks; known for its conversational fluency.
The Custom Instructions feature allows users to fine-tune and personalize the AI’s behavior and responses based on their needs or organizational requirements. By providing specific instructions, you can guide the AI to respond in a way that aligns with your team’s communication style or specific tasks. This feature ensures that the AI delivers more relevant, efficient, and customized outputs tailored to the context of your nonprofit.
Click the “Custom Instructions” Button: At the bottom right of the chat window, next to the “Prompts” button, you will find the Custom Instructions button. Clicking this opens the settings for creating or editing custom instructions.
Custom instructions allow you to set detailed preferences for how the AI interacts with you. Here is how to create them:
General Guidelines: You can provide overall guidelines on how the AI should respond. For example, you can ask the AI to:
Be concise in its answers.
Use formal language for donor communication.
Always summarize action items at the end of discussions.
Editing Instructions: You can revisit and modify custom instructions at any time. Simply click the Custom Instructions button again and make any necessary changes to the details.
Using Predefined Instructions: You can also create predefined instructions for different teams or departments in your organization. This is helpful when different teams have different expectations from the AI (e.g., the development team versus the communications team).
Task-Specific Instructions: You can customize instructions for specific tasks. For example, if the AI is being used to help with grant writing, you might instruct it to always use persuasive language or highlight the impact of the nonprofit’s work.
Personas: Create AI personas to align with specific roles. For example, if the AI is assisting your communications team, you can create a persona that focuses on social media engagement, while a separate persona can be tailored for donor management.
Saving Instructions: Once you have detailed your custom instructions, save them, and when enabled, the AI will apply these preferences during your chat. These instructions can be updated or revised as needed to suit different projects or workflows.
