Playground
The GreenNode Playground is a web-based interface that allows you to test and compare AI models quickly and intuitively—without writing a single line of code. Designed for both developers and AI enthusiasts, it supports a wide range of capabilities—such as chat, completions, image generation, speech processing, embeddings, and document reranking—making it a flexible and powerful environment for rapid prototyping and exploration.
Explore Various Specialized Playgrounds
GreenNode AI Platform offers seven distinct playgrounds, each tailored for different types of model interactions:
- Chat Playground: Engage with chat-based models in a conversational interface. Use system prompts to steer the model’s tone, behavior, or role.
- Language Playground: Generate and refine text using large language models in a collaborative text editor. Ideal for creative writing, instruction-following tasks, and few-shot prompting.
- Image Playground: Turn your ideas into visuals. Generate images from text or edit existing images using models like Stable Diffusion and other leading open-source models.
- Speech to Text Playground: Transcribe audio recordings into text with high-accuracy speech recognition models.
- Text to Speech Playground: Convert written text into natural-sounding speech, with configurable voice styles and pacing.
- Embedding Playground: Produce vector embeddings for text inputs—ideal for semantic search, clustering, and similarity tasks.
- Rerank Playground: Reorder and score document lists or search results based on relevance to a given query or context.
Instructions
Follow these steps to begin exploring models in the GreenNode Playground:
- Visit: https://aiplatform.console.greennode.ai/playground
- At the top, select the model type depending on your task: Chat, Completion, Image Generation, Speech, Embedding, Rerank, ...
- On the right side, pick a model from the list and configure its parameters (such as temperature, max tokens, etc).
- In the System prompt section, define instructions or context for the model. You can also include few-shot examples to guide its behavior.
- Start chatting! Enter prompts and observe how different models respond.
Once you're happy with a model configuration, you can save it for future use:
- Copy as code: Click <> API in the center panel to extract your current setup and chat history for use in your application.
Model parameters
When configuring model behavior in the GreenNode Playground, you can fine-tune the following inference parameters from the left-side panel:
Temperature
A floating-point value that controls the randomness of the output.
- 0: Deterministic and consistent results.
- < 1: Produces focused, accurate responses—ideal for summarization or factual Q&A.
- ≈ 1: Encourages creativity and variety in the output.
Top-p (Nucleus Sampling)
A probability threshold (between 0 and 1) that limits token selection to the smallest group whose cumulative probability exceeds the threshold.
- Helps produce fluent, natural responses while maintaining diversity.
- Example: top_p = 0.9 considers only tokens within the top 90% of probability mass.
Presence Penalty
A numeric value that penalizes tokens already present in the generated text.
- Positive values encourage the model to introduce new topics by reducing token repetition.
- Helps promote topic diversity and reduce redundancy in responses.
Max Tokens
The maximum number of tokens the model will generate. Lower values result in faster response times but shorter outputs. Adjust to balance output length and performance.
Logprobs (API only)
Specifies how many top token log-probabilities to return at each generation step.
- Useful for advanced usage such as token-level analysis, debugging, or custom sampling.
Note
For more parameters supported by the GreenNode MaaS API, refer to the vLLM documentation and OpenAI documentation
Compare Mode
Run two model setups side by side for easy comparison.
- Click Compare to open dual chat windows.
- Submit inputs to both models simultaneously.
Related Articles
Overview
GreenNode offers Model as a Service (MaaS) to help developers and businesses integrate powerful AI capabilities into their applications with ease. Whether you're working with language models, vision models, or multi-modal AI systems, GreenNode ...
