Skip to main content
Kaino.dev
Discover
Evals
News
Academics
Kaino.dev

Discover, evaluate, and compare AI tools, models, and agents.

Explore

  • Discover
  • Evaluations
  • News
  • Academics

Community

  • Twitter
  • YouTube
  • Instagram
Privacy PolicyTerms of Service

© 2026 Kaino.dev. All rights reserved.

Version 1.1.0
AionLabs: Aion 3.5 · Discover · Kaino
Discover/MODELS/AionLabs: Aion 3.5
AionLabs: Aion 3.5 logo

MODELS

AionLabs: Aion 3.5

by Aion Labs

modelaion-labs
Visit WebsiteDocumentationGitHub

Overview

A GLM-family multi-model system for roleplaying and storytelling, available through Aion Labs’ OpenAI-compatible API.

Details

AionLabs: Aion 3.5 (`aion-labs/aion-3.5`) is a text-to-text model offered through Aion Labs’ OpenAI-compatible API. The official model catalog describes GLM-family multi-model generation and lists a 256K context window with up to 32K output tokens; the API reference specifies 262,144 context tokens and 32,768 maximum completion tokens. It supports OpenAI-compatible chat and responses endpoints, streaming, tool calls, and required reasoning. Aion Labs also identifies it as an unmoderated model.

When to Use

Use for roleplaying and storytelling applications. Use when integrating a text model through OpenAI-compatible chat or responses APIs. Use when streaming output or tool calling is needed.

Getting Started

  1. Review the Aion Labs Quickstart and API reference.
  2. Configure an OpenAI-compatible client for the Aion Labs API.
  3. Send a chat-completions or responses request using `aion-labs/aion-3.5`.
  4. Use the public models endpoint to inspect available models and model metadata.

Key Features

  • •GLM-family multi-model generation.
  • •262
  • •144-token context window.
  • •Up to 32
  • •768 completion tokens.
  • •OpenAI-compatible chat and responses endpoints.
  • •Streaming and tool-call support.
  • •Required reasoning and unmoderated model status.

Capabilities

  • •text-to-text generation
  • •roleplaying
  • •storytelling
  • •chat completions
  • •responses API
  • •streaming
  • •tool calling
  • •model discovery

Last updated Sep 24, 2026