Hacktoberfest 2026: los issues que los mantenedores marcaron para octubre, abiertos y aptos para principiantes. Explorar issues de Hacktoberfest

azure-ai-projects: gpt-6-luna returns 500 on project endpoint, works on resource endpoint

Abierto
  • ¿Algo no está bien?
#49,169 3 comentarios 4 reacciones 0 asignados Ver en GitHub

Los mantenedores suelen responder en 1 día

Nadie ha tomado este issue todavía.

Evaluación

Dificultad
4/5
Tiempo estimado
3-5 días
Aptitud para principiantes
45/100
Tipo de issue
Error
Claridad
Bastante claro
Estado de actividad
Activo
Stack tecnológico
azure, python
Área
api, cloud

Línea de trabajo

Start by comparing the project endpoint with the resource endpoint using the supplied Python reproduction, then read the related issue #48943 and microsoft/agent-framework#8718. Confirm whether the project endpoint is returning incorrect model capabilities for gpt-6-luna; the expected result is successful responses with reasoning parameters supported, matching the resource endpoint. The report indicates this may be a service-side issue rather than an SDK change.

Escrito por el modelo de indexación a partir del texto del issue.

Descripción

AI Projects Service Attention
  • Package Name: azure-ai-projects
  • Package Version: 2.6.1
  • Operating System: macOS
  • Python Version: 3.12

Describe the bug
gpt-6-luna (2026-09-22, Global Standard, Sweden Central) fails on the Foundry project endpoint and works on the resource endpoint. gpt-6-sol and gpt-5.6-luna in the same project work on both. This reproduces with raw HTTP, so it looks like a service issue, not an SDK one. It's the same pattern as #48943, which was gpt-6-astra at its launch.

To Reproduce

from azure.ai.projects import AIProjectClient
from azure.identity import DefaultAzureCredential

project = AIProjectClient(
    endpoint="https://<resource>.services.ai.azure.com/api/projects/<project>",
    credential=DefaultAzureCredential(),
)
client = project.get_openai_client()

client.responses.create(model="gpt-6-luna", input="hello")
# -> 500 server_error, every time

client.responses.create(model="gpt-6-luna", input="hello", reasoning={"effort": "medium"})
# -> 400 "Unsupported parameter: 'reasoning.effort' is not supported with this model."

client.responses.create(model="gpt-6-sol", input="hello")
# -> 200

The same gpt-6-luna calls to https://<resource>.services.ai.azure.com/openai/v1/responses return 200, and reasoning effort is accepted there.

What it looks like
On a streaming call to the project endpoint, the first response.created event shows gpt-6-luna with every reasoning field empty and top_p: 1.0. gpt-6-sol shows effort medium, context all_turns and top_p: 0.98. So the project endpoint seems to treat gpt-6-luna as a model without reasoning support.

Request IDs

  • 500: 542dd8ff-9577-469e-90dc-e01e23c001da, 515f279b-6158-4138-97cc-d1c3df8cd42b
  • 400: d52a9067-b2d1-413b-b0e1-8c69cd14e804

Expected behavior
gpt-6-luna works on the project endpoint, like gpt-6-sol does.

Related: microsoft/agent-framework#8718, which shows the same Luna-only reasoning.effort rejection through Foundry Agents.

Lenguaje dominante
Python
Estrellas
5.6k
Forks
3.4k
Merge medio
1 d 18 h
PR fusionados (30 d)
202

Preparar el entorno

Abrir en Codespaces

Inicia el contenedor de desarrollo del proyecto en tu navegador, con tu propia cuenta de GitHub.

Primeros pasos

  1. Lee el issue completo y luego la guía de contribución del proyecto.
  2. Comenta en el issue que vas a ocuparte — evita que dos personas hagan lo mismo.
  3. Haz un fork del repositorio y trabaja en una rama.
  4. Abre un pull request que haga referencia al número del issue.

Más de Azure/azure-sdk-for-python

Todos los issues de Azure/azure-sdk-for-python

Issues similares

Más issues de Python

Recibe los nuevos issues en tu correo

Un resumen breve de issues de GitHub para principiantes.