Unverified50% confidenceBenchmarkExact time
Under a 4M TPM premise, long Agent turns with a 400K context can trigger at most 10 requests per minute (hitting the TPM cap), and full-context requests only 4 per minute.
1
Sources
50%
Confidence
Long-term
Relevance
10/15/2026
First Seen
Sources
Related Entities
Related Claims
UnverifiedAn agent task requiring 20 LLM calls with 200 tokens generated each time takes approximately 133 seconds to complete at 30 TPS, but only 13 seconds at 300 TPS.80% similarUnverifiedA short conversation turn consumes about 1300 Tokens/request, in which case the RPM limit (3000) is hit before the TPM limit (4 million); in long-context scenarios TPM becomes the bottleneck.75% similarVerifiedA provider's rate limits typically take effect simultaneously across three dimensions: requests per minute (RPM), tokens per minute (TPM), and tokens per day (TPD).74% similarUnverifiedIn concurrent scenarios, generation speed is evenly distributed among multiple requests; for example, at 40 token/s for a single request, each of 4 concurrent requests gets approximately 10 token/s74% similarUnverified一个Agent的总Token消耗可能达到50K-100K，而同样的对话在Chat模式下可能只需要5K72% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/1034047API
curl https://kongchang.com/api/v1/knowledge/claims/1034047MCP
get_claim(id=1034047)