Why moderate: one template-boundary shift from the group rather than an exact signature member.
AionLabs: Aion-3.0 is Aion Labs's model. Its measured tokenizer sits with the Z.ai group, which is evidence that it builds on the Z.ai tokenizer. Tokenizer reuse is normal engineering, open vocabularies travel between labs, and this says nothing against Aion Labs's authorship of the model itself.
| We checked | What we found | Where it came from |
|---|---|---|
| How it counts tokens | Counted cleanly, and no other model in the catalogue counts the same way | tk_c018f9f2 |
| How its API is set up | Set up identically to 1 other listing, field for field | api_a6b72370 |
| How much it can read and write | Reads up to 131,072 tokens at once, writes up to 32,768 | its own listing |
| Whether it thinks before answering | Always reasons, and cannot be turned off | its own listing |
| Who served our requests | AionLabs | we measured it, 3 September 2026 |
Click a row to open the full comparison.
We sent AionLabs: Aion-3.0 fifty short pieces of text and recorded what each one cost it in tokens. The bars below are those costs. Two models built on the same tokenizer produce the same bars; a model built on a different one produces a different set, which is what makes this a fingerprint.
Overhead subtracted: 7 prompt tokens.
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 each contribute...
Supported parameters
include_reasoningmax_tokensreasoningresponse_formattemperaturetool_choicetoolstop_p
defaults: {}
