refactor: drop request-body token override, keep env-only config

The unauthenticated /api/validate-model endpoint should not let callers
raise the token budget; the env var / admin setting alone fixes #883.
Also import the shared constants in the settings registry instead of
duplicating them, and move the setting to the Features group alongside
the other validation settings.
This commit is contained in:
dayuan.jiang
2026-07-11 22:30:20 +09:00
parent c1e4b1c9bf
commit eb2fa69f58
4 changed files with 35 additions and 32 deletions
+16 -11
View File
@@ -10,6 +10,11 @@
// - ADMIN_PASSWORD / SETTINGS_FILE: bootstrap values, env-only to avoid lockout
// - Per-provider reasoning/thinking tuning vars: env-only (see env.example)
import {
DEFAULT_MODEL_VALIDATION_MAX_TOKENS,
MAX_MODEL_VALIDATION_MAX_TOKENS,
} from "@/lib/model-validation"
export type SettingType = "string" | "secret" | "number" | "boolean" | "enum"
export interface SettingDef {
@@ -88,17 +93,6 @@ export const SETTINGS_REGISTRY: SettingDef[] = [
label: "Max Output Tokens",
min: 1,
},
{
key: "MODEL_VALIDATION_MAX_TOKENS",
group: "generation",
type: "number",
label: "Model Validation Max Tokens",
description:
"Token budget for model test requests. Increase for reasoning models that may emit thinking tokens first.",
min: 1,
max: 64000,
default: "1000",
},
// ── Access Control ───────────────────────────────────────────────
{
@@ -134,6 +128,17 @@ export const SETTINGS_REGISTRY: SettingDef[] = [
label: "Validation Timeout (ms)",
min: 1000,
},
{
key: "MODEL_VALIDATION_MAX_TOKENS",
group: "features",
type: "number",
label: "Model Validation Max Tokens",
description:
"Token budget for model test requests. Increase for reasoning models that may emit thinking tokens first.",
min: 1,
max: MAX_MODEL_VALIDATION_MAX_TOKENS,
default: String(DEFAULT_MODEL_VALIDATION_MAX_TOKENS),
},
{
key: "ENABLE_HISTORY_XML_REPLACE",
group: "features",