- Type
- Open weightsMIT License
- Params
- 111B
- Context
- 131K
12B active per word · about 98K words of context
Our take
Written Sep 30, 2026GLM 4.5 Air is a text model you can download and run yourself, with a licence that allows commercial use, changes and redistribution. Its measured quality is mid-field: it sits in the lower half of the Arena Text field, so treat it as a budget pick rather than a quality one.
Use it for budget-conscious chat and general text work where a mid-field model is acceptable, or when you want to run a model on a single modern graphics card. The licence allows commercial use, changes and redistribution (MIT). Skip it if you need a model near the top of the Arena boards for coding, maths or creative writing.
The case for it
- The licence allows commercial use, changes and redistribution (MIT), so the terms are not the thing that decides this one.
- Only 12 billion of its 110.5 billion parameters are active per token, so memory in use is closer to a small model than to a mid-size one.
- The cheapest listed offer sits well under the vendor's own rate for both input and output, so the host you pick changes the bill more than it changes the model.
The case against it
- Mid-field on measured quality: 112th of 168 on Arena Text (overall) as of 25 Sep 2026, and 114th of 168 on Arena Creative Writing as of 25 Sep 2026 — human preference votes, not correctness.
- Coding and maths are not its strong suit: 111th of 168 on Arena Coding as of 25 Sep 2026 and 106th of 163 on Arena Maths as of 25 Sep 2026, both preference-based placings rather than pass rates.
How good is it?
EverydayGeneral questions and everyday reasoning
Arena Text (overall)112th of 168 · 1373
CodingWriting and fixing code on its own
Arena Coding111th of 168 · 1427
Arena Coding is the only board that has scored it for this.
AgenticPlanning, calling tools, staying on task
Not yet scored on Arena Agent.
WritingDrafting and rewriting prose
Arena Creative Writing114th of 168 · 1327
Arena Creative Writing is the only board that has scored it for this.
These tests check whether a model follows instructions — a precondition for all the work above, but not a measure of how well that work is done.
Every published score for this model6 scoresEvery figure we hold, from 6 boards, with who ran it and a link to the source — including the boards no rating above is built on.
Can you run it yourself?
GeForce RTX 4090 · 24 GB
Too large for this card. The weights do not fit even with part of them offloaded to system memory.
Apple M1 Pro (16-core GPU) · 32 GB
Too large for this card. The weights do not fit even with part of them offloaded to system memory.
Comfortable fit
Apple M1 Ultra (64-core GPU) · 128 GB
Room to spare. 22.7 GB spare means a 10% error in the size would not change the answer.
Memory use by level
Against a 24 GB card.
What is quantisation? →This model on every device we track71 devicesThe Q4 build most people download, on each device: what the weights come to, how much context the memory leaves, and whether it runs. Smallest device that runs it first.
Check against your own machine → · Where to rent it hosted →
Or rent it from someone else
Prices checked 4 hours ago — each listing carries its own date.
- per 1M tokens
- $0.13 in / $0.85 out
- Context served
- 131K
- Throughput
- Not measured
| Provider | In / out per 1M tokens | Context | Throughput | Trains on prompts | Logs prompts | Zero retention |
|---|---|---|---|---|---|---|
| OpenRouterOpenRouter's own listing | $0.13 / $0.85checked 4 hours ago | 131K | not measured | Unknown | Unknown | Unknown |
| Novita AIbf16Direct and through OpenRouter | $0.13 / $0.85checked 4 hours ago | 131K98K max reply through OpenRouter | 51 tok/sthrough OpenRouter | DirectUnknownThrough OpenRouterNo | DirectUnknownThrough OpenRouterNo | DirectUnknownThrough OpenRouterConfirmed |
| SiliconFlowfp8Through OpenRouter | $0.14 / $0.86checked 4 hours ago | 131K118K max reply | 34 tok/s | No | No | Confirmed |
| Z.AIfp8Through OpenRouter | $0.20 / $1.10checked 4 hours ago | 131K98K max reply | 27 tok/s | No | No | Confirmed |
Novita AIAcross the 4 listings we hold: 3 say they do not train on prompts (1 of them only through OpenRouter), 0 say they do and 1 does not say. 3 appear in the zero-retention registry we check (1 of them only through OpenRouter); the rest are unknown to us.
What each host's API supports
From the parameter list each endpoint publishes. Streaming is omitted: nothing we hold reports it, for any model.
| Provider | Tool calling | JSON output | Strict schema |
|---|---|---|---|
| OpenRouterOpenRouter's own listing | ✓ | ✗ | ✗ |
| Novita AIbf16Direct and through OpenRouter | ✓ | ✗ | ✗ |
| SiliconFlowfp8Through OpenRouter | ✓ | ✗ | ✗ |
| Z.AIfp8Through OpenRouter | ✓ | ✗ | ✗ |
Tool calling: 4 of 4 listings say yes. JSON output: 0 of 4 listings say yes, 4 say no. Strict schema: 0 of 4 listings say yes, 4 say no.
When we formed this view
Recent changes
What moved
first indexed by our pipelineEach date is the day we first saw the change, or the day the maker announced it.
What we do not know about this model yet
- We hold no measured file for it, so all 3 sizes on this page are calculated from the parameter count.
- Nothing we hold says whether an endpoint streams, so we do not show it either way.
- 1 of 4 listings does not say whether it trains on prompts, and 1 answers only through OpenRouter, not for its own listing.
- We hold no batch or off-peak rate for any of its listings.
- We hold a decode speed for it, but no prompt-processing (prefill) figure, so how long the input side of a job takes is unknown to us.
Licence and identifiers
What the licence allowsMIT License, what it allows commercially, and the identifiers you need to pull this model — its Hugging Face repo, our slug and a machine-readable card.
Licence
MIT License
Fully permissive: do anything with attribution. No patent grant, unlike Apache-2.0.
Identifiers
- Hugging Face
- zai-org/GLM-4.5-Air
- Architecture
- Mixture of experts
- Takes in, gives back
- Text in, text out
- Catalogue slug
- z-ai-glm-4-5-air