| 模型名称 | 速率限制 | 概述 |
| gemma2 | Unlimited | 27b本地模型，无审查，隐私 |
| Llama3-70B-8K | RPM = 120，TPM = 120000 | |
| Llama3-8B-8K | RPM = 240，TPM = 240000 | 快速推理 |
| Mistral-7B-32K | RPM = 60，TPM = 60000 | |
| ERNIE-Speed-128K | RPM = 60，TPM = 300000 | 128k长上下文，高性能 |
| ERNIE-Speed-8K | RPM = 300，TPM = 300000 | |
| ERNIE-Lite-8K | RPM = 300，TPM = 300000 | |
| ERNIE-Tiny-8K | RPM = 300，TPM = 300000 | 快速推理 |
| Qwen1.5-0.5B | RPM = 1500，下列总日限10k | 高速率，无审查 |
| Qwen1.5-1.8B | RPM = 720 | 高速率，无审查 |
| Qwen1.5-7B | RPM = 300 | 无审查 |
| Qwen1.5-14B | RPM = 150 | 无审查 |
| Sqlcoder-7B | RPM = 300 | SQL指令模型，无审查 |
| DeepSeek-Coder-6.7B | RPM = 300 | 代码模型，无审查 |
| DeepSeek-Math-7b | RPM = 300 | 数学模型，无审查 |
| Llama3-8B-cf | RPM = 300 | 无审查 |
| Spark-Lite | QPS = 2 (RPM <= 120) | 联网实时查询 |
| Hunyuan-lite-256K | QPS = 5 (RPM <= 300) | 高性能，256K超长上下文 |
暂无评论