Aion-3.0 is a multi-model roleplaying and storytelling system from AionLabs, built on the GLM family of models.
It uses a collaborative generation process in which multiple specialized models contribute to a response, with an emphasis on narrative structure and tension.
| Developer | AionLabs |
|---|---|
| Model ID | aion-labs/aion-3.0 |
| Release Date | July 6, 2026 |
| Context Window | 131,072 tokens (~131K) |
| Max Output Tokens | 32,768 tokens |
| Modalities | Image, Text → Text |
| Capabilities | file input, parallel tool calls, reasoning, streaming, structured (JSON schema) output, tool/function calling, vision, web search |
Model identity
The exact model id is aion-labs/aion-3.0, and it is presented as Aion-3.0 from AionLabs.
Architecture description
Public listings describe it as a multi-model collaborative generation system rather than a single model response generator.
Family
It is described as being built on the GLM family of models.
Supported features
Public listings for this model mention reasoning and tool use/function calling support.
Availability
It is exposed through OpenAI-compatible APIs on several provider listings.
Roleplaying
The model is described as being tuned for roleplaying-style generation.
Storytelling
The model is described as being tuned for storytelling and narrative-focused output.
Creative writing
Its collaborative generation design is presented as useful for producing stronger narrative structure.
Aion-3.0 Mini
AionLabs also lists Aion-3.0 Mini as a related model, described as using the DeepSeek family instead of GLM.
Sources: https://developer.puter.com/ai/aion-labs/aion-3.0/ · https://www.aionlabs.ai/ · https://venice.ai/models/aion-labs-aion-3-0
Measured on the AI/ML API gateway across live traffic. Figures are medians unless stated. Last updated 2026-10-12.
Aion-3.0 is documented as suited to interactive fiction, character-driven chat, and other creative-writing tools where narrative depth and thematic nuance matter more than raw speed. It is also described as a good fit for roleplaying and storytelling use, with stronger narrative structure, tension, and conflict in mature or darker themes.
It is suitable for collaborative storytelling and creative writing assistance, especially when building long-form narratives or scenes that depend on subtext and emotional turns. It is also positioned for interactive roleplaying scenarios with consistent character personas and extended-session memory for plot and detail continuity.
import requests r = requests.post( "https://api.aimlapi.com/v1/chat/completions", headers={"Authorization": "Bearer " + AIMLAPI_KEY}, json={ "model": "aion-labs/aion-3.0", "messages": [ { "role": "user", "content": "Hello!" } ] }, ) print(r.json())
const r = await fetch("https://api.aimlapi.com/v1/chat/completions", { method: "POST", headers: { Authorization: `Bearer ${process.env.AIMLAPI_KEY}`, "Content-Type": "application/json", }, body: JSON.stringify({ "model": "aion-labs/aion-3.0", "messages": [ { "role": "user", "content": "Hello!" } ] }), }); console.log(await r.json());
curl -X POST https://api.aimlapi.com/v1/chat/completions \ -H "Authorization: Bearer $AIMLAPI_KEY" \ -H "Content-Type: application/json" \ -d '{"model":"aion-labs/aion-3.0","messages":[{"role":"user","content":"Hello!"}]}'
OpenAI-compatible — swap the base URL and it works with your existing SDK.
| Type | Price |
|---|---|
| Input | |
| Output | |
| Cached input |
Aion-3.0 has separate cache-read pricing at $0.75 per 1M tokens, and one source also shows a free tier for use through Puter.js with a user-pays option for developers.
| Model | Input | Output | Context | Best for |
|---|---|---|---|---|
Aion-3.0 This page | Agentic workflows and structured output | |||
| Reasoning + agents | ||||
| Reasoning + agents |
Aion-3.0 has a 131,072 tokens context window and can return up to 32,768 tokens.
Aion-3.0 takes image, text as input and returns text.
Use aion-labs/aion-3.0 as the model id. Requests go to https://api.aimlapi.com/v1/chat/completions.
Aion-3.0 became available on July 7, 2026.
Aion-3.0 is priced at input $4.1262 / 1M tokens, output $8.2524 / 1M tokens, cached input $1.03155 / 1M tokens.
Yes, Aion-3.0 can stream responses as they are generated.
Yes, Aion-3.0 accepts image input alongside text.
Yes, it supports tool use, including parallel tool calls.
Yes, web search is listed among its features and capabilities.
