零延遲代理與多模型 SDK 規範
開發者 SDK 與 LiteLLM 路由代理接入指南
提供面向 LiteLLM Proxy、Vercel AI SDK 以及 OpenAI 官方 SDK 的一鍵配置程式碼，快速落地自適應預算風控與智慧分流。
LiteLLM Proxy Config (`config.yaml`)
model_list:
# Cost-optimized intelligent router configuration
- model_name: cost-router-fast
litellm_params:
model: deepseek/deepseek-chat
api_base: https://api.deepseek.com
api_key: os.environ/DEEPSEEK_API_KEY
- model_name: cost-router-reasoning
litellm_params:
model: deepseek/deepseek-reasoner
api_base: https://api.deepseek.com
api_key: os.environ/DEEPSEEK_API_KEY
- model_name: cost-router-flagship
litellm_params:
model: anthropic/claude-3-5-sonnet-20241022
api_key: os.environ/ANTHROPIC_API_KEY
router_settings:
routing_strategy: cost-based-routing
max_budget_per_req: 0.005
fallbacks: [{"cost-router-fast": ["cost-router-flagship"]}]Vercel AI SDK (TypeScript)
import { createOpenAI } from '@ai-sdk/openai';
import { generateText } from 'ai';
// TokenCosts AI Dynamic Routing Gateway
const openrouter = createOpenAI({
baseURL: 'https://openrouter.ai/api/v1',
apiKey: process.env.OPENROUTER_API_KEY,
headers: {
'HTTP-Referer': 'https://tokencostsai.com',
'X-Title': 'TokenCosts AI-Router',
},
});
export async function executeCostOptimizedPrompt(prompt: string, isComplex = false) {
const targetModel = isComplex
? 'deepseek/deepseek-r1'
: 'deepseek/deepseek-chat';
return await generateText({
model: openrouter(targetModel),
prompt,
});
}GitHub README Dynamic Cost Badges
Embed transparent real-time API cost badges directly in your open-source repository documentation:
<!-- TokenCosts AI Real-time Cost Badge -->
[](https://tokencostsai.com/models/deepseek-v3)