GLM-4.5 Air
Language modelActive
GLM-4.5 Air
Z.ai
- Released
- July 20, 2025
- Data date
- September 8, 2026
- Speed
- 65.4 tokens/sOutput throughput
- Price
- Input · Output / 1M tokens
- Input
- Text
- Output
- Text
GLM-4.5 Air carries the hybrid-reasoning direction of GLM-4.5 into a smaller MoE class with 106B total and 12B active parameters. The context remains 131,072 tokens.
Z.ai publishes Air as a separate text checkpoint with open weights and also makes it available through APIs. Air is therefore the compact text-oriented delivery within the GLM-4.5 family.
Specifications and access
| Specification | Value and source |
|---|---|
| Weights | Open weightsSource |
| Access | API, LocalSource |
| Parameters | 106BSource |
| Active parameters | 12BSource |
| Context window | 131,072 tokensSource |
| License | MITSource |
| Architecture | MoE hybrid reasoning modelSource |
| Parameters | 106 BSource |
| Active parameters | 12 BSource |
| Input | TextSource |
| Output | TextSource |
| Reasoning mode | thinking, non-thinkingSource |
| Tool calling | YesSource |
| API model ID | glm-4.5-airSource |
Measurements without matching peer values
These measurements have no matching peer values under the same test conditions. Their original values and sources remain available here.
GLM-4.5-Air aggregate score (Show measurement, test conditions, and source)
- Source value
- 59.8
- Score
- 59.8
- Metric
- points
- Unit
- points
- Category
- composite
- Direction
- Higher is better
- Source type
- vendor-reported
- Evaluator
- GLM-4.5-Air model card
- Status
- active
- Retrieved at
- 2026-09-08
- Methodology
- Vendor-reported result. 12-industry-benchmark composite; exact benchmark weighting not exposed
Artificial Analysis Intelligence Index (Show measurement, test conditions, and source)
- Source value
- 11.0889137659922
- Score
- 11.0889137659922
- Metric
- points
- Unit
- points
- Direction
- Higher is better
- Source type
- independent-evaluator
- Evaluator
- Artificial Analysis
- Status
- active
- Retrieved at
- 2026-09-08
- Methodology
- Independent evaluation. unknown
AA-Omniscience Index (Show measurement, test conditions, and source)
- Source value
- -61.5333333333333
- Score
- -61.5333333333333
- Metric
- points
- Unit
- points
- Direction
- Higher is better
- Source type
- independent-evaluator
- Evaluator
- Artificial Analysis
- Status
- active
- Retrieved at
- 2026-09-08
- Methodology
- Independent evaluation. unknown
Output speed (Show measurement, test conditions, and source)
- Source value
- 65.3921265021372
- Score
- 65.3921265021372
- Metric
- tokens/s
- Unit
- tokens/s
- Direction
- Higher is better
- Source type
- independent-evaluator
- Evaluator
- Artificial Analysis
- Status
- active
- Retrieved at
- 2026-09-08
- Methodology
- Independent evaluation. unknown; default 10K-input workload only
- Context
- Artificial Analysis performance timing; output speed is decoded after the first streaming chunk. Provider basis: serving-provider median. Deprecated model. Artificial Analysis only continues benchmarking the default 10K-input workload.
Latency reasoning time (Show measurement, test conditions, and source)
- Source value
- 30.584721846209334
- Score
- 30.584721846209334
- Metric
- seconds
- Unit
- seconds
- Direction
- Lower is better
- Source type
- independent-evaluator
- Evaluator
- Artificial Analysis
- Status
- active
- Retrieved at
- 2026-09-08
- Methodology
- Independent evaluation. unknown; default 10K-input workload only
- Context
- Artificial Analysis performance timing; output speed is decoded after the first streaming chunk. Provider basis: serving-provider median. Deprecated model. Artificial Analysis only continues benchmarking the default 10K-input workload.
Latency input time (Show measurement, test conditions, and source)
- Source value
- 2.64740640100001
- Score
- 2.64740640100001
- Metric
- seconds
- Unit
- seconds
- Direction
- Lower is better
- Source type
- independent-evaluator
- Evaluator
- Artificial Analysis
- Status
- active
- Retrieved at
- 2026-09-08
- Methodology
- Independent evaluation. unknown; default 10K-input workload only
- Context
- Artificial Analysis performance timing; output speed is decoded after the first streaming chunk. Provider basis: serving-provider median. Deprecated model. Artificial Analysis only continues benchmarking the default 10K-input workload.
Answer generation time (end-to-end timing breakdown) (Show measurement, test conditions, and source)
- Source value
- 7.646180461552333
- Score
- 7.646180461552333
- Metric
- seconds
- Unit
- seconds
- Direction
- Lower is better
- Source type
- independent-evaluator
- Evaluator
- Artificial Analysis
- Status
- active
- Retrieved at
- 2026-09-08
- Methodology
- Independent evaluation. unknown; default 10K-input workload only
- Context
- Artificial Analysis performance timing; output speed is decoded after the first streaming chunk. Provider basis: serving-provider median. Deprecated model. Artificial Analysis only continues benchmarking the default 10K-input workload.
End-to-end reasoning time component (Show measurement, test conditions, and source)
- Source value
- 30.584721846209334
- Score
- 30.584721846209334
- Metric
- seconds
- Unit
- seconds
- Direction
- Lower is better
- Source type
- independent-evaluator
- Evaluator
- Artificial Analysis
- Status
- active
- Retrieved at
- 2026-09-08
- Methodology
- Independent evaluation. unknown; default 10K-input workload only
- Context
- Artificial Analysis performance timing; output speed is decoded after the first streaming chunk. Provider basis: serving-provider median. Deprecated model. Artificial Analysis only continues benchmarking the default 10K-input workload.
End-to-end input time component (Show measurement, test conditions, and source)
- Source value
- 2.64740640100001
- Score
- 2.64740640100001
- Metric
- seconds
- Unit
- seconds
- Direction
- Lower is better
- Source type
- independent-evaluator
- Evaluator
- Artificial Analysis
- Status
- active
- Retrieved at
- 2026-09-08
- Methodology
- Independent evaluation. unknown; default 10K-input workload only
- Context
- Artificial Analysis performance timing; output speed is decoded after the first streaming chunk. Provider basis: serving-provider median. Deprecated model. Artificial Analysis only continues benchmarking the default 10K-input workload.
Our tests
SVG short test
One SVG prompt and fixed output settings. Captures are shown only with 0 reported reasoning tokens or documented lack of reasoning controls. Identical test conditions do not make the models' internal compute identical.
Earlier tests
Pelican on a bicycle
A community classic for free-form SVG drawing.
GLM-4.5 Air
Reasoning (requested): high
Current attempt
- Run date
- Oct 1, 2026
- API provider
- openrouter
- Requested API model
- z-ai/glm-4.5-air
- API model reported in response
- z-ai/glm-4.5-air
- Reasoning setting
- high
- Pinned endpoint
- Z.AI | z-ai/glm-4.5-air
- Temperature
- Provider default
- Token limit including reasoning
- 8,192 tokens
- Test protocol
- community-visual-tests-v1
- Technical checks
- Passed, not a quality rating
- Response time
- 30.7 s
Visually reviewed. This individual test does not establish overall model quality.
Prompt and test conditions
Generate an SVG of a pelican riding a bicycle
Token limit including reasoning: 8,192 tokens
Task origin (Simon Willison)Popularity
52.8 / 100
Sources and data date
| Type | Evidence and data date |
|---|---|
| Speed | Median across serving providers. Output speed after the first streaming chunk, not total response time. Classified as deprecated by Artificial Analysis. Artificial Analysis maintains only the default 10K-input workload there. Retrieved September 8, 2026. |
| Price | Base API prices without caching or batch discounts. Higher context tiers and other rates are listed under Costs. |
| Input | Documented modalities. Retrieved September 8, 2026. |
| Output | Documented modalities. Retrieved September 8, 2026. |
| Research status | Research date September 8, 2026. 4 source URLs checked. This documents the inspected sources, not an exhaustive inventory of every publication. |
| Unresolved | detailed-benchmarks Original wording Card says 12 benchmark evaluation but exposes only the aggregate score; detailed benchmark graphic/table not available in checked raw card. |
| Unresolved | release-date Original wording No normalized release-date field. |
| Additional source | Model metadata · GLM-4.5 Air model card · Editorial description reviewed October 2, 2026 |
| Additional source | Model metadata (retrieved September 8, 2026) · Z.ai GLM-4.5 documentation · Editorial description reviewed October 2, 2026 |
| Additional source | OpenRouter: Popularity index (derived), CC BY 4.0 |
| Additional source | Vercel AI Gateway: Popularity index (derived), CC BY 4.0 |
| Additional source | Creative Commons Attribution 4.0 |