📃Text Block

Text blocks generate text based on a text prompt.

Title

• Click the pencil icon next to the block’s title to edit it. This is also the block’s variable name, which can be used to dynamically populate data in other blocks.

Prompt

• Write a prompt that guides the AI to generate your desired text.

  • Use variables within your prompt to dynamically add data from other blocks.

System Prompt

A system prompt for an LLM is a set of instructions or context given to the model at the beginning of an interaction to guide its behavior, set parameters for the conversation, or provide background information for its responses. The default system prompt is "You are a helpful assistant." However, you can define a more specific system prompt for your needs, such as "You are simulating a most unhinged Wojak meme creator ..."

Variables

Variables can be used to dynamically populate data supplied by other blocks. Click on a variable name from the list to add it to anywhere in your prompt. Variable names match a block’s title.

Text Output

Use the slider to amp up the creativity of the text output from “Predictable” to “Insane.”

Model

Select which text generation model you’d like your glif to use.

The text models that we currently support (as of 12/10/24) include:

Chat GPT-3.5 Turbo
Chat GPT-4
Chat GPT-4o
Claude 3 - Haiku
Claude 3 - Opus
Claude 3 - Sonnet
Claude 3.5 - Haiku
Claude 3.5 - Sonnet
DeepSeek Coder
Gemini Flash 1.5
Gemini Pro 1.5
Gemma-7b-it
Gemma2-9b-it
Hermes 3
Llama 3
Llama-3.2-3b-instruct
Llama-3.3-70b-instruct
Llama3 Sonar 70B Online
Llama3 Sonar 8B Online
Llama3-70b-8192 on groq
Llama3-8b-8192 on groq
Llama3.1 Sonar 70B Online
Llama3.1 Sonar 8B Online
Mistral Large
Mixtral-8x7b-32768
Sonar Medium Online
WizardLM-2-8x22b

Advanced Text Block Controls

Max Tokens

Tokens are the individual units of text that the model processes. One token equals approximately four characters in the English language. More tokens = more characters in the output.

Seed

The "seed" is the starting number used to generate random outputs. It serves as a specific starting point for the randomization process. Identifying a specific seed will produce similar final results for each run.