TextGenerator
Gives access to a large language model for text generation.
| Memory category | Instances |
|---|
Member index 5
Description
A TextGenerator instance lets you use a large language model (LLM) to
generate text based on a system prompt from you and a user prompt from the
player. The most common use of the API is for creating interactive non-player
characters (NPCs).
For example, in a survival experience, your system prompt for a talking animal
might be
"You are a very busy beaver. You end all statements by mentioning how you need to get back to work on your dam.".
Users could ask the beaver about water in the area, the size of a nearby
forest, predators, etc.
The novelty of LLM responses can help create unique, delightful moments for players, but using the API effectively requires a bit of creativity and tuning. System prompts can be very extensive, so don't hesitate to include a long string with lots of detail.
Rate limits
Requests are initially limited to 100 per minute, which scales up based on the number of concurrent users.
History 6
- 690 Add TopP
- 690 Add Temperature
- 690 Add Seed
- 688 Add GenerateTextAsync
- 688 Add SystemPrompt
- 688 Add TextGenerator
Members 5
GenerateTextAsync
| Parameters (1) | ||
|---|---|---|
| request | Dictionary | |
| Returns (1) | ||
| Dictionary | ||
This method returns text generated by an LLM based on the provided system and user prompts, as well as any other optional paramaters that have been set.
The request argument for this method should be a dictionary with the
following structure:
| Key Name | Data Type | Description | Required |
|---|---|---|---|
UserPrompt | string | Optional prompt from the user that initiates the chat. This could be a question, statement, or command that the user wants the model to respond to. | No |
ContextToken | string | Prompt history context token containing a summarization of the previous prompt requests and responses in a conversation up to the current request. If no token is provided, a new token is generated and returned in the response. Providing a previously generated context token restores the conversation state into the current request. | No |
MaxTokens | number | The maximum number of tokens in the response generated by the model. Expected to be an integer of at least 1. This limits the length of the response, preventing overly long or incomplete answers. Non-integral numbers are rounded to the nearest integer. | No |
This method returns a dictionary with the following structure:
| Key Name | Data Type | Description |
|---|---|---|
GeneratedText | string | The generated response. |
ContextToken | string | A token containing the summarization of a previously passed context token and the current generated response. This token can be passed into subsequent requests to maintain the state of the current conversation. Subsequent requests generate new tokens with updated conversation state. Extracting the token and providing it maintains the ongoing conversation context. |
Model | string | The model and version that generated the response. |
| Thread safety | Unsafe |
|---|
History 1
- 688 Add GenerateTextAsync
Seed
| Type | Default | |
|---|---|---|
| int | 0 | |
Sets a fixed seed for the random number generator, allowing reproducible
responses in cases where the same input parameters are used across
multiple requests. By setting the same seed value, you can obtain
identical results for debugging, testing, or evaluation purposes. The
value of Seed should be an integer. Non-integral values will be
truncated. Default is 0.
| Thread safety | ReadSafe |
|---|---|
| Category | Data |
| Loaded/Saved | true |
SystemPrompt
| Type | Default | |
|---|---|---|
| string | ||
Provides context to the model about its role, tone, or behavior during the
conversation. This parameter can guide the model on how to respond,
setting expectations like "You are an assistant" or
"Use a formal tone".
| Thread safety | ReadSafe |
|---|---|
| Category | Data |
| Loaded/Saved | true |
History 1
- 688 Add SystemPrompt
Temperature
| Type | Default | |
|---|---|---|
| float | 0.699999988 | |
Controls the "creativity" or randomness of the model's responses. Values
closer to 1 increase randomness, while values closer to 0 make the
responses more focused and deterministic. Values outside the accepted
range are clamped to the range of [0.4, 1.0]. Default is 0.7.
| Thread safety | ReadSafe |
|---|---|
| Category | Data |
| Loaded/Saved | true |
History 1
- 690 Add Temperature
TopP
| Type | Default | |
|---|---|---|
| float | 0.899999976 | |
Helps the model narrow or expand the range of possible words to sample
from while generating the next token. This setting narrows the token
choices to only contain words that together make up a certain percentage
of total likelihood (for example, 90%). A lower TopP means the model
sticks to closer and more predictable choices, while a higher TopP opens
the door to more diverse and creative responses. Values outside the
accepted range are clamped to the range of [0.5, 1.0]. Default is 0.9.
| Thread safety | ReadSafe |
|---|---|
| Category | Data |
| Loaded/Saved | true |