| Spec | GLM 5.3 | GLM 5.2 |
|---|---|---|
| Lowest input | $0.20Wafer | $0.19Wafer |
| Lowest output | $1.80Wafer | $1.80DeepInfra |
| Lowest cached input | $0.125DeepInfra | $0.105DeepInfra |
| Context | 1.05M | 1.05M |
| Max output | 524K | 524K |
| Providers | 12 | 8 |
| Released | August 14, 2026 | June 16, 2026 |
| Weights | Open | Open |
| Lab | Z.ai | Z.ai |
| Model ID | z-ai/glm-5.3 | z-ai/glm-5.2 |
USD per 1M tokens, as of November 5, 2026.
| Provider | GLM 5.3 | GLM 5.2 |
|---|---|---|
| DeepInfra | $0.5625 / $2.50 | $0.5625 / $1.80 |
| Doubleword | $1.40 / $4.40 | $0.93 / $3.00 |
| Inference.net | $0.90 / $3.00 | $1.30 / $4.40 |
| Morph | $1.19 / $6.00 | $1.40 / $4.40 |
| Novita AI | $1.40 / $4.40 | $1.40 / $4.40 |
| StreamLake | $1.40 / $4.40 | $1.40 / $4.40 |
| Wafer | $0.20 / $1.80 | $0.19 / $4.40 |
| Chutes | – | $1.25 / $3.95 |
| Mistral AI | $1.40 / $4.40 | – |
| Nebius | $1.40 / $4.40 | – |
| Phala | $1.40 / $4.40 | – |
| RunInfra | $1.30 / $4.20 | – |
| Sail Research | $0.98 / $3.08 | – |
Input / output, USD per 1M tokens.
Input: GLM 5.2 at $0.19 per million input tokens vs $0.20 (GLM 5.3), 5% lower. Output: the same lowest output price, $1.80. Lowest listed prices as of November 5, 2026.
Both list 1.05M tokens of context.
7 providers serve both: Morph, DeepInfra, Novita AI, Inference.net, StreamLake, Doubleword, Wafer. GLM 5.3 has 12 providers in total and GLM 5.2 has 8.
GLM 5.3 from $0.125 per million cached input tokens (DeepInfra). GLM 5.2 from $0.105 (DeepInfra).
Yes. Both use the same OpenAI-compatible endpoint, https://api.routerplus.com/v1, and key. Change the model field from "z-ai/glm-5.3" to "z-ai/glm-5.2".
