Verified65% confidenceFactTime unknown
Taking GPT-4o as an example, the JSON Schema description of each tool consumes an average of 200–500 tokens, and when the system integrates 50 tools, the tool description section alone consumes 10,000 to 25,000 tokens.
3
Sources
65%
Confidence
Long-term
Relevance
5/31/2026
First Seen
Sources
Related Entities
Related Claims
UnverifiedThe JSON Schema description of each tool ranges from tens to hundreds of tokens, and when the number of tools exceeds 30–50, the tool definitions themselves may consume thousands of tokens of context.78% similarUnverifiedThe complete definitions of 200 tools of medium complexity average about 30,000-80,000 Tokens.67% similarUnverifiedA single detailed tool description may consume hundreds of tokens, and ten tools can easily exceed a fixed overhead of one thousand tokens.65% similarUnverified在一个典型项目查询中发送了45,000 tokens的上下文，但真正有用的只有约5,000 tokens62% similarUnverifiedIn a typical scenario of 20 rounds of interaction, if the system prompt occupies 15,000 tokens and each round produces an average of 2,000 tokens, by the 20th round the input token count for a single request balloons to 55,000 tokens.59% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/3592API
curl https://kongchang.com/api/v1/knowledge/claims/3592MCP
get_claim(id=3592)