Z.ai models: prices and the hardware each needs
19 models, 5 sold per token by 15 vendors. Vendors are in alphabetical order; the price is the cheapest list price among them, per million tokens.
| Model | Size | Sold per token by | From, input | Output | GPU memory |
|---|---|---|---|---|---|
| codegeex4-all-9b | 9.4 B | not sold per token | 23.5 GB | ||
| GLM 5 | 754 B | AWS, Azure | $1.00 | $3.20 | 1884.7 GB |
| GLM 5.1 | 754 B | Alibaba Cloud, Azure, Huawei Cloud, Nebius | $0.809 | $3.23 | 1884.7 GB |
| GLM 5.2 | 753 B | Alibaba Cloud, Azure, Cloudflare, Core42, Databricks, DeepInfra, Huawei Cloud, Nebius, Telnyx, Together AI | $0.5625 | $1.80 | 1883.3 GB |
| GLM 5.3 | 753 B | Alibaba Cloud, Azure, Cloudflare, Crusoe, Databricks, DeepInfra, Fireworks AI, Huawei Cloud, Mistral, Snowflake, Telnyx, Together AI | $0.5625 | $2.50 | 941.7 GB |
| GLM 5.3 Flash | 321 B | Azure, Cloudflare, Databricks, DeepInfra, Fireworks AI, Telnyx, Together AI | $0.135 | $0.45 | 401.6 GB |
| GLM-4-9B-0414 | 9.4 B | not sold per token | 23.5 GB | ||
| GLM-4.5 | 358.34 B | not sold per token | 895.8 GB | ||
| GLM-4.5-Air | 110.47 B | not sold per token | 276.2 GB | ||
| GLM-4.5-Air-FP8 | 110.47 B | not sold per token | 138.1 GB | ||
| GLM-4.6 | 356.79 B | not sold per token | 892 GB | ||
| GLM-4.7 | 358.34 B | not sold per token | 895.8 GB | ||
| GLM-4.7-Flash (zai-org) | 31.22 B | not sold per token | 78 GB | ||
| GLM-4.7-FP8 | 358.34 B | not sold per token | 447.9 GB | ||
| GLM-5-FP8 | 753.86 B | not sold per token | 942.3 GB | ||
| GLM-5.1-FP8 | 753.86 B | not sold per token | 942.3 GB | ||
| GLM-5.2-FP8 | 753.33 B | not sold per token | 941.7 GB | ||
| GLM-5.3-BF16 | 753.33 B | not sold per token | 1883.3 GB | ||
| GLM-Z1-32B-0414 | 32.57 B | not sold per token | 81.4 GB |