Konstantinos Thermos 28ffc814c0 fix(llm): retry OpenAI import with max_completion_tokens ...

Newer OpenAI models (gpt-5.x) reject max_tokens with a 400 and demand
max_completion_tokens, so AI booking import failed and no reservations
were created. Send max_tokens first (understood by the classic chat API
and every local server), and only on a 400 naming max_completion_tokens
resend the whole request with that parameter instead — mirroring the
existing json_schema->json_object fallback beside it. Detection is by
API response, not a model-name allowlist.

Fixes #1760

2026-08-05 15:13:17 +02:00

..

clients.test.ts

fix(llm): retry OpenAI import with max_completion_tokens

2026-08-05 15:13:17 +02:00

extraction-router.test.ts

3.2.0 (#1426)

2026-07-05 01:13:04 +02:00

lenient-json.test.ts

fix(llm-parse): tolerate non-strict JSON from Gemini and similar models

2026-07-24 12:42:38 +02:00

llm-client.factory.test.ts

3.2.0 (#1426)

2026-07-05 01:13:04 +02:00

llm-config.resolver.test.ts

refactor(server): extract addon enablement into the addons domain

2026-07-28 20:22:01 +02:00

llm-local.service.test.ts

3.3.0 (#1520)

2026-07-11 22:28:32 +02:00

llm-parse.service.test.ts

chore: move settings backend to proper DI

2026-07-26 22:23:57 +02:00

llm-prompt.test.ts

3.2.0 (#1426)

2026-07-05 01:13:04 +02:00

mime-email.test.ts

fix(llm-parse): decode .eml as MIME before extracting text (#1724)

2026-07-29 15:14:21 +02:00

nuextract.test.ts

3.2.0 (#1426)

2026-07-05 01:13:04 +02:00

ollama-format.client.test.ts

3.3.0 (#1520)

2026-07-11 22:28:32 +02:00

text-extract.test.ts

fix(llm-parse): decode .eml as MIME before extracting text (#1724)

2026-07-29 15:14:21 +02:00