wehub-resource-sync
29 lines
1.4 KiB
TypeScript
29 lines
1.4 KiB
TypeScript
import { thinkingContext } from '@/lib/ai/thinking-context';
|
|
import type { ThinkingConfig } from '@/lib/types/provider';
|
|
|
|
/**
|
|
* PBL v2 runtime LLM calls force-disable thinking.
|
|
*
|
|
* The instructing turn forces `begin_turn` via `tool_choice`, which several
|
|
* providers reject when thinking is on — DeepSeek returns 400 "Thinking mode
|
|
* does not support this tool_choice". We never intentionally enabled thinking
|
|
* on these turns (the PBL v2 client sends no `thinkingConfig`); some pinned
|
|
* models just default it on. Disabling it removes the incompatibility without
|
|
* losing any behavior we relied on.
|
|
*
|
|
* For OpenAI-compatible providers (e.g. DeepSeek) thinking is injected by the
|
|
* fetch wrapper in `providers.ts`, which reads the per-request config from the
|
|
* `thinkingContext` AsyncLocalStorage. The agents call the AI SDK directly
|
|
* (not via `callLLM`/`streamLLM`), so nothing seeds that store — we do it here.
|
|
* The SDK call must be started INSIDE `run` (its consumption can be outside) so
|
|
* the lazily-issued fetch inherits the context, matching `streamLLM`.
|
|
*
|
|
* Scope: PBL v2 runtime only (instructor / evaluator). Generation (planner) is
|
|
* intentionally untouched.
|
|
*/
|
|
const PBL_V2_THINKING_DISABLED: ThinkingConfig = { mode: 'disabled', enabled: false };
|
|
|
|
export function withThinkingDisabled<T>(startCall: () => T): T {
|
|
return thinkingContext.run(PBL_V2_THINKING_DISABLED, startCall);
|
|
}
|