| Spec | Qwen3.8 Max | GLM 5.3 |
|---|---|---|
| Lowest input | $1.65DeepInfra | $0.20Wafer |
| Lowest output | $4.951DeepInfra | $1.80Wafer |
| Lowest cached input | $0.206DeepInfra | $0.125DeepInfra |
| Context | 256K | 1.05M |
| Max output | 131K | 524K |
| Providers | 1 | 12 |
| Released | September 3, 2026 | August 14, 2026 |
| Weights | Proprietary | Open |
| Lab | Qwen | Z.ai |
| Model ID | qwen/qwen3.8-max | z-ai/glm-5.3 |
USD per 1M tokens, as of November 5, 2026.
| Provider | Qwen3.8 Max | GLM 5.3 |
|---|---|---|
| DeepInfra | $1.65 / $4.951 | $0.5625 / $2.50 |
| Doubleword | – | $1.40 / $4.40 |
| Inference.net | – | $0.90 / $3.00 |
| Mistral AI | – | $1.40 / $4.40 |
| Morph | – | $1.19 / $6.00 |
| Nebius | – | $1.40 / $4.40 |
| Novita AI | – | $1.40 / $4.40 |
| Phala | – | $1.40 / $4.40 |
| RunInfra | – | $1.30 / $4.20 |
| Sail Research | – | $0.98 / $3.08 |
| StreamLake | – | $1.40 / $4.40 |
| Wafer | – | $0.20 / $1.80 |
Input / output, USD per 1M tokens.
| Model | 1,000 chat replies2K in, 500 out each | 1,000 agent tasks50K in, 5K out each | Monthly batch job500M in, 50M out |
|---|---|---|---|
| Qwen3.8 MaxDeepInfra | $5.78 | $107.26 | $1,072.55 |
| GLM 5.3Wafer | $1.30 | $19.00 | $190.00 |
On each model's cheapest provider by input plus output list price.
Input: GLM 5.3 at $0.20 per million input tokens vs $1.65 (Qwen3.8 Max), 8.2x lower. Output: GLM 5.3 at $1.80 per million output tokens vs $4.951 (Qwen3.8 Max), 2.8x lower. Lowest listed prices as of November 5, 2026.
GLM 5.3, with 1.05M tokens vs 256K.
1 provider serves both: DeepInfra. Qwen3.8 Max has 1 provider in total and GLM 5.3 has 12.
Qwen3.8 Max from $0.206 per million cached input tokens (DeepInfra). GLM 5.3 from $0.125 (DeepInfra).
Yes. Both use the same OpenAI-compatible endpoint, https://api.routerplus.com/v1, and key. Change the model field from "qwen/qwen3.8-max" to "z-ai/glm-5.3".
