Usage Log Query

Query usage logs for a specific Token using the API Key directly — no login required. Useful for usage analysis, cost accounting, and reconciliation.

Endpoint

GET /api/log/token

Request Parameters

Parameter	Type	Required	Description
key	string	Yes	API Key, e.g. sk-xxx
p	int	Yes	Page number, starting from 1
page_size	int	Yes	Items per page, max 100
start_timestamp	int64	Yes	Start time (Unix seconds)
end_timestamp	int64	Yes	End time (Unix seconds)
type	int	No	Log type: 0 all, 1 topup, 2 consume, 5 error, 6 refund
model_name	string	No	Model name (fuzzy match)

Example

cURL

# Query consumption logs from the last 7 days, page 1, 20 items per page
curl "https://api.ezmodel.cloud/api/log/token?\
key=sk-your-api-key&\
p=1&\
page_size=20&\
type=2&\
start_timestamp=1710000000&\
end_timestamp=1710604800"

Python

import requests
import time

base_url = "https://api.ezmodel.cloud/"
now = int(time.time())
week_ago = now - 7 * 86400

resp = requests.get(f"{base_url}/api/log/token", params={
    "key": "sk-your-api-key",
    "p": 1,
    "page_size": 20,
    "type": 2,  # consumption logs
    "start_timestamp": week_ago,
    "end_timestamp": now,
})
data = resp.json()
print(f"Total: {data['data']['total']} logs")
for log in data["data"]["items"]:
    print(f"  {log['model_name']}: {log['prompt_tokens']}+{log['completion_tokens']} tokens")

Response Format

{
  "success": true,
  "message": "",
  "data": {
    "page": 1,
    "page_size": 20,
    "total": 156,
    "items": [
      {
        "id": 42,
        "created_at": 1710500000,
        "type": 2,
        "username": "your_username",
        "token_name": "my-token",
        "model_name": "gpt-4o",
        "quota": 5000,
        "prompt_tokens": 1200,
        "completion_tokens": 800,
        "use_time": 3,
        "is_stream": true,
        "group": "default",
        "other": "{\"model_ratio\":1.5,\"request_path\":\"/v1/chat/completions\"}"
      }
    ]
  }
}

Response Fields

Field	Type	Description
id	int	Log ID
created_at	int64	Creation time (Unix seconds)
type	int	Log type
token_name	string	Token name
model_name	string	Model name
quota	int	Quota consumed
prompt_tokens	int	Input token count
completion_tokens	int	Output token count
use_time	int	Request duration (seconds)
is_stream	bool	Whether streaming was used
group	string	User group
other	string	Extended info (JSON string with pricing ratios, cache details, etc.)

Error Handling

Error Message	Cause	Solution
key is required	API Key not provided	Pass the full API Key in the key parameter
record not found	Invalid or non-existent API Key	Verify the Key is correct
start_timestamp and end_timestamp are required	Missing time range	Add both timestamp parameters
p and page_size are required	Missing pagination parameters	Add p and page_size parameters

Best Practices

Performance Tips

1. Use a reasonable time range — narrower ranges yield faster queries
2. Use reasonable page sizes — 20-50 recommended, max 100
3. Filter by model if you only need usage for specific models
4. Paginate through all data — increment p until items is empty or page * page_size >= total

Next Steps

• See Getting Started for basic API usage
• See Code Examples for integration in various languages