GLM 4.5 Air

Tool callingReasoning

Hugging Face Launch post Website

Lighter GLM-4.5 variant for fast coding assistance and cheaper agents

Context: 131K
Released: 1y ago
Parameters: —
Input / 1M tokens: $0.13
Output / 1M tokens: $0.85
Cached input / 1M tokens: $0.02

Sources:models.devOpenRouter

Benchmarks

DeepSWE is the primary coding benchmark where an exact model evaluation exists. Other sourced evaluations appear below; missing scores are not zero.

Not evaluated in DeepSWE 1.1 as of 2026-09-22. Results are only shown for exact model versions in the official dataset. Official leaderboard.

No benchmark results have been matched to this model yet.

Specifications

Developer: Z.ai
Model family: GLM
Type: Language model
Predecessor: —
Successor: —
Released: July 28, 2025
Model ID: z-ai/glm-4.5-air
Openness: Open weights
License: —
Parameters: —
Input modalities: text
Output modalities: text
Modality: text->text
Tokenizer: Other
Max output tokens: 98K
Knowledge cutoff: 2025-04
Hugging Face: zai-org/GLM-4.5-Air
Source freshness: Metadata synced Oct 8, 2026, 10:17 PM UTC

Additional source evidence

Source model: zhipuai/glm-4.5-air
Source family: glm-air
Last updated: 2025-07-28
Knowledge cutoff: 2025-04
Model context: 131K
Model output limit: 98K
Open weights: Yes
File attachments: No
Reasoning: Yes
Tool calling: Yes
Temperature control: Yes

Downloadable weights

Hugging Face

Published benchmarks

Benchmark	Score	Evaluation details	Evidence
Artificial Analysis Coding Index	23.8 · index	2026-05-30	Source
SciCode	30.6 · percent correct	2026-05-30	Source
Terminal-Bench Hard	20.5 · success rate	2026-05-30	Source

Complete source record (JSON)

{
  "id": "zhipuai/glm-4.5-air",
  "name": "GLM-4.5-Air",
  "description": "Lighter GLM-4.5 variant for fast coding assistance and cheaper agents",
  "family": "glm-air",
  "last_updated": "2025-07-28",
  "knowledge": "2025-04",
  "attachment": false,
  "reasoning": true,
  "tool_call": true,
  "temperature": true,
  "modalities": {
    "input": [
      "text"
    ],
    "output": [
      "text"
    ]
  },
  "limit": {
    "context": 131072,
    "output": 98304
  },
  "benchmarks": [
    {
      "name": "Artificial Analysis Coding Index",
      "score": 23.8,
      "source": "https://openrouter.ai/z-ai/glm-4.5-air/benchmarks",
      "metric": "index",
      "date": "2026-05-30"
    },
    {
      "name": "SciCode",
      "score": 30.6,
      "source": "https://openrouter.ai/z-ai/glm-4.5-air/benchmarks",
      "metric": "percent correct",
      "date": "2026-05-30"
    },
    {
      "name": "Terminal-Bench Hard",
      "score": 20.5,
      "source": "https://openrouter.ai/z-ai/glm-4.5-air/benchmarks",
      "metric": "success rate",
      "date": "2026-05-30"
    }
  ],
  "release_date": "2025-07-28",
  "open_weights": true,
  "weights": [
    {
      "label": "Hugging Face",
      "url": "https://huggingface.co/zai-org/GLM-4.5-Air"
    }
  ]
}

Infrastructure

Provider availability

All providers

Provider	Endpoint	Context	Input /M tokens	Output /M tokens	Cached input /M tokens	Cache write /M tokens	Uptime 1d	Data handling
Novita	novita/bf16	131K	$0.13	$0.85	$0.02	—	100.00%	ZDR: Yes Training: Unknown Prompt logging: Unknown Retention: Unknown ZDR source Privacy policy Checked 2026-10-15
SiliconFlow	siliconflow/fp8	131K	$0.14	$0.86	—	—	99.50%	ZDR: Yes Training: Unknown Prompt logging: Unknown Retention: Unknown ZDR source Privacy policy Checked 2026-10-15
Z.AI	z-ai/fp8	131K	$0.20	$1.10	$0.03	—	99.41%	ZDR: Yes Training: Unknown Prompt logging: Unknown Retention: Unknown ZDR source Privacy policy Checked 2026-10-15

Provider data:OpenRouter

Deployment graph

Can I run this?

Vertical Tabs only shows compatibility when it has structured evidence tying this exact model to hardware, runtime, quantization, memory use, and observed performance.

Browse hardware