The agent that grows with you
Python aiai-agentai-agentsanthropicchatgptclaudeclaude-codecodexhermeshermes-agentllmnous-researchopenai
/** Hermes' reasoning levels, in ascending order — mirrors the backend's
* VALID_REASONING_EFFORTS (hermes_constants.py). `none` is not a level: it's
* thinking disabled. */
export const REASONING_EFFORTS = ['minimal', 'low', 'medium', 'high', 'xhigh', 'max', 'ultra'] as const
export type ReasoningEffort = (typeof REASONING_EFFORTS)[number]
/** The scale plus the off state — the full set a config value may hold. */
export const REASONING_EFFORT_VALUES = ['none', ...REASONING_EFFORTS] as const
export type ReasoningEffortValue = (typeof REASONING_EFFORT_VALUES)[number]
/** Hermes' built-in level when neither the surface nor the profile config
* specifies one (mirrors the backend's own fallback). */
export const DEFAULT_REASONING_EFFORT: ReasoningEffort = 'medium'
/** True for a real level (case-insensitive, trimmed); `none` is not a level. */
export const isReasoningEffort = (value: string): value is ReasoningEffort =>
REASONING_EFFORTS.includes(value.trim().toLowerCase() as ReasoningEffort)