Side-by-side comparison · models
GLM 4.5 vs GLM 4.5 Air
What we have checked about each, line by line. Every value shows where it came from and when we last confirmed it.
Models
GLM 4.5
GLM-4.5 is our latest flagship foundation model, purpose-built for agent-based applications.
Models
GLM 4.5 Air
GLM-4.5-Air is the lightweight variant of our latest flagship model family, also purpose-built for agent-centric applications.
Checked facts
Line by line
| Measure | GLM 4.5 | GLM 4.5 Air |
|---|---|---|
| Input price | $0.600 per million tokens | $0.200 per million tokens |
| Output price | $2.20 per million tokens | $1.10 per million tokens |
| Price on z-ai | $0.110 per million cached input tokens | $0.030 per million cached input tokens |
| Price on z-ai | $0.600 per million input tokens | $0.200 per million input tokens |
| Price on z-ai | $2.20 per million output tokens | $1.10 per million output tokens |
| Price on novita | Not checked yet | $0.025 per million cached input tokens |
| Price on novita | Not checked yet | $0.130 per million input tokens |
| Price on novita | Not checked yet | $0.850 per million output tokens |
| Price on siliconflow | Not checked yet | $0.140 per million input tokens |
| Price on siliconflow | Not checked yet | $0.860 per million output tokens |
| Latest version | z-ai/glm-4.5 | z-ai/glm-4.5-air |
| Time to first token (ms) on z-ai | 4,552 | 982 |
| Output speed (tokens a second) on z-ai | 25 | 31 |
| Time to first token (ms) | Not checked yet | 2,390 |
| Time to first token (ms) on novita | Not checked yet | 720 |
| Time to first token (ms) on siliconflow | Not checked yet | 1,085 |
| Output speed (tokens a second) | Not checked yet | 82.8 |
| Output speed (tokens a second) on novita | Not checked yet | 54 |
| Output speed (tokens a second) on siliconflow | Not checked yet | 53 |
| Tokens served on OpenRouter in a day | 217.7M (for 10 Nov 2026) | 430.8M (for 10 Nov 2026) |
| Tool calls on OpenRouter in a day | 1.2K (for 10 Nov 2026) | 5.8K (for 10 Nov 2026) |
| Artificial Analysis Intelligence Index | 12.8 | 11.1 |
| LMArena text rating (Elo) | 1,430 | 1,384 |
| Design Arena rating (Elo) | 1,166 | 1,184 |
| Lech mazur writing | 73.4% (for 3 Aug 2025) | Not checked yet |
| Weirdml | 40.6% (for 5 Aug 2025) | Not checked yet |
| Gso bench | Not checked yet | 2.9% (for 20 July 2025) |
A line appears only when at least one record has a value we confirmed from a source. Values we have not confirmed are left out, never guessed. Records of different types share fewer lines.
Add a third
Compare with one more record
How this works
Share it, or build your own
This page works for any two or three records in the directory, of any type. The link stays the same whichever order you name them in, so you can share it as it is. Change the records in the address, for example /directories/vs/claude-code-vs-cursor, or start from our editorial comparisons.