Unverified50% confidenceSolutionExact time
If you simply use an ordinary prompt to have Argon write a 60,000-word novel, it will stop quickly; after explicitly writing in the system prompt that "the output budget for this response is 1 million tokens, and you must not summarize, skip, or stop early," the detail in the city tiles became more than six times richer.
1
Sources
50%
Confidence
Long-term
Relevance
10/14/2026
First Seen
Sources
Related Entities
Related Claims
UnverifiedArgon bills thinking tokens as output tokens, so the high level significantly raises the per-call cost even on simple questions.60% similarUnverifiedBy default, Argon does not run to the full 1 million tokens, and in tests it often stops at 32K or 65K tokens; through the interactions API, when the limit is reached the server returns a continuation token, and the client can carry that token plus a higher budget to continue generation from the breakpoint, or enable background mode to let the server automatically chain 65K segments together.59% similarUnverifiedA Bilibili content creator used Trae Code to complete the requirements for a publishing tool and consumed over 500 credits in a single day, equivalent to the total amount earned from five days of free daily check-ins56% similarUnverifiedStuffing an entire codebase into every request incurs real latency and cost penalties; one million tokens is a capacity ceiling, not a recommendation.56% similarUnverifiedOpenClaw's token consumption is extraordinarily high; a 10-yuan quota can be burned through in minutes when invoking tools and browsing websites, and a morning or a full day of coding work can consume several million tokens.54% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/1030351API
curl https://kongchang.com/api/v1/knowledge/claims/1030351MCP
get_claim(id=1030351)