GLM-4.5-Air
Lighter GLM-4.5 variant for fast coding assistance and cheaper agents
Shogo Agent Index
UnrankedNot ranked yet: 1 agentic benchmark measured, and we need at least 2 plus an overall capability score. It will be ranked as results are published.
Agentic coding
20
11% measured
Long-horizon autonomy
30
estimated
Knowledge-work agents
18
estimated
Computer use
9
estimated
Security & context learning
20
estimated
Agentic benchmarks
| Benchmark | Score | Normalised | Index weight | Harness / source |
|---|---|---|---|---|
| SWE-bench Verified | est. | 8% | predicted 54 | |
| Terminal-Bench 2.0 | est. | 8% | predicted 12 | |
| DeepSWE | est. | 6% | predicted 12 | |
| FrontierSWE | est. | 5% | predicted 1 | |
| CursorBench | est. | 4% | predicted 19 | |
| GSO-Bench | 2.9% | 4% | OpenHands | |
| METR Time Horizons | est. | 10% | predicted 54 | |
| Vending-Bench 2 | est. | 6% | predicted 3 | |
| PostTrainBench | est. | 4% | predicted 12 | |
| APEX-Agents | est. | 8% | predicted 13 | |
| GDPval | est. | 7% | predicted 20 | |
| Remote Labor Index | est. | 5% | predicted 0 | |
| DeepResearch Bench | est. | 5% | predicted 42 | |
| OSWorld 2.0 | est. | 6% | predicted 1 | |
| OSWorld | est. | 4% | predicted 21 | |
| Cybench | est. | 5% | predicted 26 | |
| CL-bench | est. | 5% | predicted 13 |
Other benchmarks
| Benchmark | Category | Score | Source |
|---|---|---|---|
| Humanity's Last Exam | Reasoning | 8.1% | source |
API hosts
Live prices and uptime for every host serving GLM-4.5-Air via OpenRouter.
| Host | Input | Output | Cache read | Agent $/M | Uptime 24h | Quant |
|---|---|---|---|---|---|---|
| Novita | $0.13 | $0.85 | $0.025 | $0.080 | 100.0% | bf16 |
| Z.AI | $0.20 | $1.10 | $0.030 | $0.11 | 100.0% | fp8 |
| SiliconFlow | $0.14 | $0.86 | — | $0.17 | 100.0% | fp8 |