User Guide
- Manage Prompts
- Create Prompt
- Edit Version
- Generate Version
- Rename Version
- Delete Version
- Set Primary Version
- Lock Version
- Compare Versions
- Token Count and Cost
- Update Prompt Use Case
- Copy Prompt to Workspace
- Move Prompt to Workspace
- Test Prompts
- Prompt Chat
- Response Alignment
- Optimize Prompts
- Prompt Scoring
- Scoring Criteria
- Prompt Heatmap
- Prompt Balance
- Prompt Risk
- Workspaces
- Add Workspace
- Rename Workspace
- Delete Workspace
- Switch Workspace
- Projects
- Add Project
- Delete Project
- Update Project Brief
- Assign Prompt to Project
- API Clients
- Add API Client
- Edit API Client
- Delete API Client
- Regenerate API Client
- Require API Client
- Connect AI Providers
- AI21 Labs
- Amazon Bedrock
- Anthropic
- Azure OpenAI
- Azure Serverless
- Cohere
- Deep Infra
- Fireworks AI
- Google Gemini
- Groq
- Hugging Face
- Hyperbolic
- IBM WatsonX
- Inception
- Meta
- Mistral
- Moonshot AI
- OpenAI
- Perplexity
- Together AI
- xAI
Prompt Chat
On this page you'll find the following:
Get Started
You can chat with each version of your prompts as part of your testing process to ensure you get the proper expected results. Using Zatomic's built-in chat functionality allows you to continuously fine-tune your prompts quickly and easily to verify they produce the desired outcomes.
To get started, go to the version's Chat tab. Enter your query or question and click the purple Go button.
Tip: Use many different queries and input variations to thoroughly cover all aspects of your prompt.
The AI response will start streaming and then complete after a couple moments.
Tip: You can have as many user inputs and AI responses as needed for a given chat; there is no limit.
Reset the Chat
You can reset the chat session for a prompt version by clicking the Reset link at the bottom-right of the user input box; this will bring up the Reset Chat modal box. Confirming to reset the chat will remove all user inputs and AI responses from the prompt version.
Chat with Different Models
To change which AI model is used to chat with the prompt, click the Settings link near the top-right of the page.
This will open a side panel where you can select different models and change the temperature setting. Make your model selection, click the green Apply button, then enter a query or question in the user input box.
Tip: Changing the temperature allows you to control how creative the response will be based on the prompt. Values closer to 0 will be less creative while values closer to 1 will be more creative. This can help fine-tune your prompt.
The temperature setting isn't valid for the following models: GPT-5, GPT-5 Mini, GPT-5 Nano. If one of these models is chosen, the temperature setting will be hidden.
Tip: You can change models in between responses.
