Complete Qianwen / QwenCloud branding and gateway upgrade

Abierto
  • ¿Algo no está bien?
#669 0 comentarios 0 reacciones 0 asignados Ver en GitHub

Los mantenedores suelen responder en 1 día

Nadie ha tomado este issue todavía.

Evaluación

Dificultad
5/5
Tiempo estimado
Más de una semana
Aptitud para principiantes
25/100
Tipo de issue
Nueva funcionalidad
Claridad
Bastante claro
Estado de actividad
Activo
Stack tecnológico
go
Área
backend-api-design

Línea de trabajo

Start with core/relay/adaptor/qwencloud/adaptor.go and the related FastGPT discussion at issue #7503 to locate the channel setup, default URL resolution, and help text described here. Trace how new and existing channels choose their base URL and how the adaptor composes request paths. Done means regional defaults and branding are consistent, legacy and explicit URLs still work, and tests cover resolution and request composition.

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

Descripción

Summary

Complete the user-facing brand upgrade from Bailian / Alibaba Model Studio to 千问 AI 平台 (Qianwen AI Platform) for mainland China and QwenCloud internationally. Align newly created chat channels with the new gateways while keeping old channel IDs, gateways, and configurations usable.

Motivation

The QwenCloud adaptor currently returns:

https://dashscope-intl.aliyuncs.com/compatible-mode/v1

Its mainland-China guidance also points to https://dashscope.aliyuncs.com/compatible-mode/v1.

Source: qwencloud/adaptor.go, commit c8af6371.

FastGPT consumes model/provider metadata through its plugin system and uses AIProxy for the applicable channel routing. Updating a FastGPT display name alone does not update this default.

Proposed Solution

Upgrade the existing channel setup and help text, with explicit regional selection and these defaults for newly created pay-as-you-go chat configurations:

Platform / region OpenAI-compatible Base URL
QwenCloud / international, Singapore https://ap-southeast-1.maas.qwencloudapi.com/compatible-mode/v1
Qianwen AI Platform / mainland China https://maas.qianwenaiapi.com/compatible-mode/v1
  • Upgrade remaining legacy-brand labels, help text, and corresponding FastGPT/plugin setup to the regional new brands. Do not retain the old brand as a parallel primary option for new channels.
  • Align new-channel setup with the new gateway. If an existing channel with an empty base URL would be silently redirected by changing DefaultBaseURL(), preserve its prior behavior through versioned/default-resolution compatibility rather than requiring users to migrate.
  • Preserve explicit channel URLs, old channel IDs/adaptor keys, and legacy runtime routing. The legacy endpoints remain usable; a branding update must not break old channels or their identification.
  • Reuse current configuration interactions, distinguish domestic and international selection, and update official API-key/documentation links.
  • Verify chat request URL composition using the adaptor's existing transport.
  • Scope the first PR to chat configuration. The adaptor also advertises other modes; validate those separately rather than treating a chat endpoint list as proof of support for every mode.

Acceptance Criteria

  • Existing channel branding and setup are upgraded consistently across the affected components; new channels use the correct regional gateway.
  • Existing channels, including empty/default URL cases, have documented compatibility behavior.
  • Explicit base URLs, old channel IDs, and legacy calls remain functional despite the user-facing brand change.
  • Tests cover default resolution, explicit overrides, and request-path composition.
  • Help text and any corresponding FastGPT/plugin metadata agree with the implementation.
  • The PR identifies the AIProxy release and any FastGPT/plugin dependency update needed to deliver it.

Alternatives Considered

Administrators can configure the base URL manually, but that leaves the brand upgrade incomplete. Updating the existing entry and new-channel setup, with a compatible resolver for old channels, addresses both the new-user experience and continuity for existing users.

Additional Context

Related FastGPT discussion: https://github.com/labring/FastGPT/issues/7503.

References: Qianwen AI Platform, QwenCloud documentation.

If you have questions about API keys, endpoint selection, or setup, please refer to the corresponding official site:

Lenguaje dominante
Go
Estrellas
570
Forks
120
Merge medio
5 h 52 min
PR fusionados (30 d)
5

Preparar el entorno

  • Incluye un Dockerfile o un archivo de Docker Compose
  • Sin plantilla de pull request
  • Sin guía de contribución

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 labring/aiproxy

Todos los issues de labring/aiproxy

Issues similares

Más issues de Go

Recibe los nuevos issues en tu correo

Un resumen breve de issues de GitHub para principiantes.