fix(ai): ask a thinking model again when it answered with reasoning alone

Ollama serves thinking models such as Gemma 4 on /v1/chat/completions with
the thought in a separate `reasoning` field. A long briefing can come back
with empty `content`: the thought used the whole token budget, or the
answer itself was parsed into `reasoning`. The background briefing always
asks to reason, so every briefing on such a model ended in the empty-reply
sentinel, which the chain logged as status 0 / cause "network" in the same
second the server logged 200, and the read path showed as a timeout.

The Local client and the OpenAI-compatible gateway now treat a reply with
reasoning and no answer as a lesson: the endpoint and model are remembered
in the reasoning dialect cache as answering only unreasoned, the call is
asked once more with reasoning_effort "none", and later calls ask that way
until the lesson expires. A test drives the real client, safeFetch and the
pinned dispatcher against a loopback server that answers the way Ollama
serves Gemma 4, and reproduces the reported log line without the fix.

Failed hops now carry what the call ran into: `failure` (timeout, aborted,
refused, reset, socket, parse, empty_response, reasoning_only, ...) and a
code-shaped `code` on ai.chain.link_failed, closed vocabulary only. The
cause only says "timeout" for a deadline, and the briefing failure marker
records the kind so the empty state suggests a longer timeout only when
time ran out.

Refs #1267
This commit is contained in:
Marc-André Bombeck committed 2026-10-22 13:01:10 +02:00
1 parent bb3ee6b855
commit 221960af88
11 files changed
+1040 -51

No files matched your search

+5
View File
@@ -64,6 +64,7 @@ import { resolveUserTimezone } from "@/lib/tz/resolver";
import { insightResultSchema, type InsightResult } from "@/lib/ai/types";
import { AI_BUDGETS, resolveInsightsMaxTokens } from "@/lib/ai/ai-budgets";
import { resolveEffectiveTimeoutMs } from "@/lib/ai/effective-timeout";
import { providerFailureKind } from "@/lib/ai/failure-kind";
import {
recordBriefingFailure,
readBriefingFailure,
@@ -855,6 +856,10 @@ export const POST = apiHandler((request: NextRequest) =>
: "provider-error",
locale,
httpStatus: failureHttpStatus,
failureKind:
e instanceof AllProvidersFailedError
? (e.attempts[0]?.failureKind ?? null)
: providerFailureKind(e).failure,
});
if (e instanceof AllProvidersFailedError) {
annotate({