MODEL CAPABILITIES
Context and token usage
What a token is for billing, where to read the counts, and the limits that apply.
Models read and write tokens. Tessen bills by tokens, so a request's cost follows from its token counts. Every response reports them:
USAGE
"usage": { "prompt_tokens": 11, "completion_tokens": 2, "total_tokens": 13 }| Field | Meaning |
|---|---|
prompt_tokens | Input: your messages, system prompt, tool definitions and tool results. |
completion_tokens | Output: the answer, tool-call arguments, and — for reasoning models — the reasoning tokens. |
total_tokens | The sum. |
Limits
max_tokenscaps output. On the commercial families the most you can ask for is 16384; above that the request is refused with 413request_too_largeand nothing is spent.- The context window (input plus output) is not published, so these docs state no number. Keep conversations short and watch
usage.prompt_tokens. - If
max_tokensis reached,finish_reasonislength.
Cost
Cost is tokens times the published price per million tokens. On Tessen's families the exact settled cost is in the final streamed frame and in the wallet ledger.
A RESPONSE WITH USAGE
{ "id": "b9f325286dab4acc903ac13dd4836fe2", "object": "chat.completion", "created": 1793282245, "model": "z1", "choices": [ { "index": 0, "message": { "role": "assistant", "content": "ok", "name": null, "tool_calls": null, "tool_call_id": null, "reasoning": null, "reasoning_content": null }, "finish_reason": "stop" } ], "usage": { "prompt_tokens": 11, "completion_tokens": 2, "total_tokens": 13 }}