Documentation
Documentation

Models

Choose a model and find its API ID, capabilities and limits.

Choose an available model

The catalog lists each model’s public ID, current availability, capabilities, context window and published rates. Choose a model marked Available before making a request. A model’s planned inclusion does not mean it can accept requests yet.

StatusMeaning
AvailableAccepts requests now, subject to your account access and request limits.
Temporarily unavailableCurrently unable to accept requests. Check status before retrying.
Coming soonPlanned for release. Inference access is not open yet.
RoadmapA longer-term target. The model or release plan may change.
Available modelPublic IDMaximum output
GLM 5.3 Flash Abliteratedzai/glm-5.3-flash-uncensored131,072 tokens
GLM 5.3 Abliteratedzai/glm-5.3-uncensored131,072 tokens

Use the public model ID

Copy the Model ID from the model page into the model field of your request. Use the complete value, including its prefix. A display name is not a substitute for the API ID.

A model family’s published capabilities can differ from the features available through Arnict. Check the Arnict model page for supported inputs and request limits.

List models with the API

GET /v1/models returns an OpenAI-compatible list of model records. Authenticate with the same key you use for chat completions. The list response identifies models; use the website catalog for detailed availability and prices.

bash
curl "https://api.arnict.com/v1/models" \
  -H "Authorization: Bearer $ARNICT_API_KEY"

Read model IDs from data[].id. The owned_by field identifies the model family. Listing a model is not a guarantee that a subsequent request will succeed, so handle temporary unavailability in your application.