tokenc - Node.js SDK

Compress LLM prompts to reduce costs and latency. 100K tokens compressed in ~85ms. Zero dependencies.

Install

npm install tokenc

Usage

import { TokenClient } from "tokenc";

const client = new TokenClient({ apiKey: "your-api-key" });

const result = await client.compressInput({
  input: "Your long prompt here...",
  model: "bear-1.2",
  aggressiveness: 0.5, // 0.1 = light, 0.5 = balanced, 0.9 = aggressive
});

console.log(result.output);             // compressed text
console.log(result.tokensSaved);        // tokens removed
console.log(result.compressionRatio);   // e.g. 1.8x

With Custom Settings

import { TokenClient, CompressionSettings } from "tokenc";

const client = new TokenClient({ apiKey: "your-api-key" });

const result = await client.compressInput({
  input: "Your text...",
  model: "bear-1.2",
  compressionSettings: new CompressionSettings({
    aggressiveness: 0.7,
    maxOutputTokens: 100,
  }),
});

Performance

Requests are gzip-compressed automatically. Reuse the TokenClient instance for connection pooling.

| Input Size | E2E Latency | Throughput | |---|---|---| | 10K tokens | 38ms | 198K tok/s | | 100K tokens | 85ms | 975K tok/s | | 1M tokens | 542ms | 1.5M tok/s |

Error Handling

import { TokenClient, AuthenticationError, RateLimitError, APIError } from "tokenc";

try {
  const result = await client.compressInput({ input: "Your text...", model: "bear-1.2" });
} catch (error) {
  if (error instanceof AuthenticationError) console.log("Invalid API key");
  else if (error instanceof RateLimitError) console.log("Rate limit exceeded");
  else if (error instanceof APIError) console.log(`API error: ${error.message}`);
}

Requirements

Node.js >= 18

Links

• Get API keys: https://thetokencompany.com
• Issues: https://github.com/TheTokenCompany/tokenc-npm-sdk/issues