Unverified50% confidenceFactExact time
Self-hosted vLLM instances do not accept a reasoning_effort value of 'none', returning a literal_error indicating that the input should be 'low', 'medium', or 'high'.
1
Sources
50%
Confidence
Medium-term (~90 days)
Relevance
10/19/2026
First Seen
Valid until: 1/17/2027
Sources
Related Entities
Related Claims
UnverifiedThe endpoint adaptation strategy that ultimately emerged is: send reasoning_effort: 'none' to standard endpoints; for vLLM, switch to chat_template_kwargs after the value is rejected; and for relay endpoints that return empty streaming responses, send no reasoning parameters at all.63% similarUnverified@ai-sdk/xai@4.0.58 fixes the issue of missing type validation for batch requests and will now reject unsupported batch request types.53% similarUnverifiedCOALESCE returns the first non-NULL value in the argument list, and is commonly used to set default values.53% similarUnverifiedThe should_strip_auth function in the requests library performs a pure string comparison of hostnames, with no DNS resolution and no 127.0.0.1/localhost alias detection.51% similarUnverifiedC标准中的未定义行为（UB）涵盖有符号整数溢出、空指针解引用、越界数组访问等操作，C标准对其结果不作任何保证50% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/1068563API
curl https://kongchang.com/api/v1/knowledge/claims/1068563MCP
get_claim(id=1068563)