Alibaba Cloud Int.: Models and Prices
Alibaba Cloud Model Studio: Qwen models plus third-party open-weight models.
Alibaba Cloud Int. served 466B tokens of other labs' open-weight models over Oct 3 to Oct 9, 2026 through a large public LLM router, #22 among providers.
Open-weight models
| Model | Family | Share of model | Tokens, 7d | Input $/M | Output $/M | Tokens/s | Uptime 3d | Context |
|---|---|---|---|---|---|---|---|---|
| Qwen3.8 Flash | Qwen | 100.0% | 417B | $0.15 | $0.47 | 60 | 100.0% | 1M |
| DeepSeek V4 Flash 0423 | DeepSeek | 10.0% | 306B | $0.15 | $0.30 | 82 | 100.0% | 1M |
| Qwen3.8 27B | Qwen | 21.5% | 124B | $0.425 | $2.55 | 47 | 98.8% | 1M |
| DeepSeek V4.1 Flash | DeepSeek | 0.3% | 97B | $0.30 | $1.20 | 68 | 97.9% | 1M |
| Kimi K3 | Kimi | 2.4% | 46B | $3.45 | $17.25 | 36 | 100.0% | 1.05M |
| DeepSeek V4 Pro 0813 | DeepSeek | 0.8% | 9B | $1.12 | $3.37 | 53 | 99.7% | 1M |
| GLM 5.3 | GLM | 0.3% | 8B | $1.19 | $3.74 | 54 | 100.0% | 1M |
| DeepSeek V4 Flash 0731 | DeepSeek | - | - | $0.352 | $1.06 | 56 | 100.0% | 1M |
| GLM 5.2 | GLM | - | - | $0.966 | $3.04 | 46 | 100.0% | 1M |
| DeepSeek V4 Pro 0423 | DeepSeek | - | - | $1.42 | $2.83 | 48 | 99.1% | 1M |
| Qwen3.8 2.4T A95B | Qwen | - | - | $2.00 | $6.00 | 30 | 100.0% | 1M |
| Kimi K2.7 Code | Kimi | - | - | $0.95 | $4.00 | 30 | 100.0% | 262K |
| Qwen3.5 397B A17B | Qwen | - | - | $0.39 | $2.34 | 53 | 99.9% | 262K |
| GLM 5.1 | GLM | - | - | $1.33 | $4.18 | 47 | 98.9% | 203K |
| Qwen3.6 27B | Qwen | - | - | $0.45 | $2.70 | 54 | 99.7% | 262K |
| Qwen3.5-35B-A3B | Qwen | - | - | $0.163 | $1.30 | 107 | 98.7% | 262K |
| Qwen3.5-27B | Qwen | - | - | $0.195 | $1.56 | 38 | 99.5% | 262K |
| Qwen3.5-122B-A10B | Qwen | - | - | $0.26 | $2.08 | 81 | 97.5% | 262K |
List prices per million tokens as of October 9, 2026. Share: of all tokens served on that model; a dash where the volume is not reported.
Company
- Website
- alibabacloud.com
- Headquarters
- Singapore
- Status page
- status.alibabacloud.com
Sources: Alibaba Cloud.
Finance your GPUs
Serving tokens on your own GPUs? Tell us the cluster, the models you serve and the contracts behind them, and we will share it with partner funders that fit.
Prefer email? hello@amcompute.com