Skip to main content

Documentation Index

Fetch the complete documentation index at: https://thetokencompany.com/docs/llms.txt

Use this file to discover all available pages before exploring further.

​
Quick Start

Get your API key by creating an account. Choose your language: 
import requests

# 1. Call our API with the prompt to compress
response = requests.post(
    "https://api.thetokencompany.com/v1/compress",
    headers={"Authorization": "Bearer YOUR_API_KEY"},
    json={
        "model": "bear-1.2",
        "input": "Your long text to compress goes here...",
        "compression_settings": {"aggressiveness": 0.1}
    }
)

# 2. Print the compressed prompt
print(response.json()["output"])

​
API Endpoint

POST https://api.thetokencompany.com/v1/compress

​
Authentication

Add your API key to the request header: 
Authorization: Bearer YOUR_API_KEY
Get your API key by creating an account.

​
Request

{
    "model": "bear-1.2",
    "input": "Your text to compress",
    "compression_settings": {
        "aggressiveness": 0.1
    }
}
ParameterTypeDescription
modelstringbear-1.2 (recommended), bear-1.1, or bear-1
inputstringText to compress
compression_settings.aggressivenessfloat0.0 to 1.0 — higher means more compression. Default: 0.5

​
Response

{
    "output": "Compressed text here",
    "output_tokens": 5,
    "original_input_tokens": 12
}
FieldDescription
outputCompressed text
output_tokensToken count after compression
original_input_tokensOriginal token count

​
Gzip Compression

We recommend gzip for every request — overhead is under 1ms and it’s up to 2.5x faster for large payloads. The Python SDK and npm package enable gzip by default. For direct API usage, add Content-Encoding: gzip and send the compressed JSON body. See the full guide and benchmarks on the Gzip Compression page. 
import gzip, json, requests

payload = json.dumps({
    "model": "bear-1.2",
    "input": your_text,
    "compression_settings": {"aggressiveness": 0.1}
}).encode()

response = requests.post(
    "https://api.thetokencompany.com/v1/compress",
    headers={
        "Authorization": "Bearer YOUR_API_KEY",
        "Content-Type": "application/json",
        "Content-Encoding": "gzip"
    },
    data=gzip.compress(payload)
)

​
Protected Tokens

Parts of the input text can be protected from compression by using the <ttc_safe> tags. See the full documentation on the Protect Text page. 
compression applied here<ttc_safe>but not here</ttc_safe>

​
Pricing

ModelPrice
bear-1.2$0.05 / 1M compressed tokens
bear-1.1$0.05 / 1M compressed tokens
bear-1$0.05 / 1M compressed tokens

​
What are compressed tokens?

Compressed tokens are the tokens that were removed during compression — the difference between your input and output token counts.
Example: If you send 500 tokens and receive 200 tokens back, you are charged for the 300 tokens that were removed. The 200 remaining tokens are free.