Token 成本試算器 (Price Calculator)
依據每次請求之 Input / Output Token 數量與請求次數，精確試算並排名各供應商總成本。
重要說明：下方輸入的 Input / Output Tokens 代表「每次請求 (Per Request)」的 Token 數量，總成本會自動乘以 Request Count（請求次數）。
Total Cost = ((Input Tokens × Input USD/1M + Output Tokens × Output USD/1M) / 1,000,000) × Request Count
常用試算情境預設:
最低總成本供應商
$1.20
DeepInfra (deepinfra/deepseek-ai/DeepSeek-V3.1)
Highest Cost Provider
$7.50
SambaNova (sambanova/DeepSeek-V3.1)
最高與最低價差
$6.30
Total workload difference
最高可節省比例
84.0%
Choosing lowest vs highest
Estimated Workload Cost by Provider (USD)
各供應商預估成本排名（僅列出具備完整 Input 與 Output 報價之方案）
若某供應商缺少 Input 或 Output 任一必要價格欄位，系統不會將其納入總成本排名，以避免不完整的成本比較。
| Rank | Provider · LiteLLM ID | Input / 1M | Output / 1M | Input Cost | Output Cost | Total Cost | vs Cheapest |
|---|---|---|---|---|---|---|---|
| #1 | DeepInfra· Standard · Global deepinfra/deepseek-ai/DeepSeek-V3.1 | $0.25 | $0.95 | $0.2500 | $0.9500 | $1.20 | Cheapest |
| #2 | Novita AI· Standard · Global novita/deepseek/deepseek-v3.1 | $0.27 | $1.00 | $0.2700 | $1.00 | $1.27 | +$0.0700 (save 5.5%) |
| #3 | Baseten· Standard · Global baseten/deepseek-ai/DeepSeek-V3.1 | $0.50 | $1.50 | $0.5000 | $1.50 | $2.00 | +$0.8000 (save 40.0%) |
| #4 | Weights & Biases (Weave)· Standard · Global wandb/deepseek-ai/DeepSeek-V3.1 | $0.55 | $1.65 | $0.5500 | $1.65 | $2.20 | +$1.00 (save 45.5%) |
| #5 | Google Vertex AI (DeepSeek)· Standard · Global vertex_ai/deepseek-ai/deepseek-v3.1-maas | $0.60 | $1.70 | $0.6000 | $1.70 | $2.30 | +$1.10 (save 47.8%) |
| #6 | Together AI· Standard · Global together_ai/deepseek-ai/DeepSeek-V3.1 | $0.60 | $1.70 | $0.6000 | $1.70 | $2.30 | +$1.10 (save 47.8%) |
| #7 | Replicate· Standard · Global replicate/deepseek-ai/deepseek-v3.1 | $0.672 | $2.016 | $0.6720 | $2.02 | $2.69 | +$1.49 (save 55.4%) |
| #8 | Azure AI Foundry· Standard · Global azure_ai/deepseek-v3.1 | $1.23 | $4.94 | $1.23 | $4.94 | $6.17 | +$4.97 (save 80.6%) |
| #9 | SambaNova· Standard · Global sambanova/DeepSeek-V3.1 | $3.00 | $4.50 | $3.00 | $4.50 | $7.50 | +$6.30 (save 84.0%) |