From f1f5fa97a909bf5f174be925028e7fa5c7cf40e1 Mon Sep 17 00:00:00 2001 From: Roland Tannous Date: Mon, 30 Mar 2026 06:22:30 +0000 Subject: [PATCH] fix gemini truncated reply (max_tokens 16->64) and suppress GeneratorExit on client disconnect --- studio/backend/routes/inference.py | 2 ++ studio/backend/tests/test_providers_api.py | 2 +- 2 files changed, 3 insertions(+), 1 deletion(-) diff --git a/studio/backend/routes/inference.py b/studio/backend/routes/inference.py index 93754efc16..f8c0713b30 100644 --- a/studio/backend/routes/inference.py +++ b/studio/backend/routes/inference.py @@ -967,6 +967,8 @@ async def _proxy_to_external_provider( ): yield f"{line}\n\n" yield "data: [DONE]\n\n" + except GeneratorExit: + pass except Exception as exc: logger.error("external_provider.stream_error", error = str(exc)) finally: diff --git a/studio/backend/tests/test_providers_api.py b/studio/backend/tests/test_providers_api.py index ee88ed0f42..10f9352ca9 100644 --- a/studio/backend/tests/test_providers_api.py +++ b/studio/backend/tests/test_providers_api.py @@ -512,7 +512,7 @@ class TestVisionInference: ], }], "stream": True, - "max_tokens": 16, + "max_tokens": 64, "provider_type": provider_type, "external_model": model, "encrypted_api_key": encrypted,