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

no models in custom provider

Abierto
  • ¿Algo no está bien?
#392 5 comentarios 0 reacciones 0 asignados Ver en GitHub

Los mantenedores suelen responder en 2 días

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
Estancado
Stack tecnológico
clojure
Área
backend-api-design

Línea de trabajo

Start by tracing how configured providers and their static models are initialized and added to db[:models], then check when model sync completes relative to chat startup. Reproduce with the supplied fetchModels: false configuration and inspect /providers-list, /doctor, and a chat request. Done when the custom model is available and the default model is shown and usable.

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

Descripción

Describe the bug

ECA 0.123.2 reads a custom provider from =config.json= correctly, but the provider is not registered as an available runtime model. As a result:

  • =/config= shows the provider and =defaultModel= correctly
  • =/providers-list= shows the provider as configured, but with =0 models=
  • =/doctor= shows empty =Default model:=
  • chat requests can fail with:
    #+begin_src text :results output
    API url not found.
    Make sure you have provider '' configured properly.
    #+end_src

This seems to affect custom OpenAI-compatible providers configured with =api: "openai-chat"= and explicit static models in config.

To Reproduce

  1. Create =~/.config/eca/config.json= with a custom provider, for example:
    #+begin_src json :results output
    {
    "$schema": "https://eca.dev/config.json",
    "defaultModel": "aitunnel/qwen3.5-397b-a17b",
    "providers": {
    "aitunnel": {
    "api": "openai-chat",
    "url": "https://api.aitunnel.ru/v1",
    "key": "REDACTED",
    "fetchModels": false,
    "models": {
    "qwen3.5-397b-a17b": {}
    }
    }
    }
    }
    #+end_src
  2. Start ECA server.
  3. Open chat and run =/config=.
  4. Run =/providers-list= and =/doctor=.
  5. Send a normal chat message.

Expected behavior

If a custom provider is present in =/config= and has explicit static models in =providers..models=, ECA should register those models as available runtime models even when =fetchModels=false=.

Expected result:

  • =/providers-list= should show at least 1 model for the custom provider
  • =/doctor= should show a non-empty =Default model:=
  • chat should use =aitunnel/qwen3.5-397b-a17b= successfully

Actual behavior

  • =/config= shows the merged config correctly, including:
    • =defaultModel "aitunnel/qwen3.5-397b-a17b"=
    • the custom provider
    • =fetchModels false=
    • the static =models= entry
  • =/providers-list= shows:
    #+begin_src text :results output
    aitunnel ✓ configured
    API Key · (config) · active
    api: openai-chat
    url: https://api.aitunnel.ru/v1
    fetchModels: false
    0 models
    #+end_src
  • =/doctor= shows:
    #+begin_src text :results output
    Default model:

Logged providers:
openai: Unknown
#+end_src

  • sending a chat message can fail with:
    #+begin_src text :results output
    API url not found.
    Make sure you have provider '' configured properly.
    #+end_src

Doctor

#+begin_src text :results output
ECA version: 0.123.2

Server cmd: /home/zoya/.emacs.d/eca/eca server

Workspaces: /home/zoya/.config/eca

Default model:

Logged providers:
openai: Unknown
#+end_src

Additional context

From source inspection, this looks like a runtime model registration / initialization issue rather than a config parsing issue:

  • =/config= proves the config is loaded correctly
  • custom providers with =api "openai-chat"= appear to be supported by runtime dispatch
  • the failure seems to happen because the custom model does not end up in =db[:models]=
  • chat/model selection then has no valid runtime model to use
  • there may also be an initialization race, because model sync is started asynchronously during startup

Relevant symptoms are consistent with:

  • static models from custom providers not being surfaced into available models
  • or chat starting before model sync completes
  • or both
Lenguaje dominante
Clojure
Estrellas
1k
Forks
82
Merge medio
3 d 39 min
PR fusionados (30 d)
15

Preparar el entorno

Este proyecto no incluye contenedor de desarrollo, Dockerfile ni guía de contribución, así que la configuración corre por tu cuenta: empieza por su README y consulta nuestra guía para la primera contribución para los pasos generales.

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 editor-code-assistant/eca

Todos los issues de editor-code-assistant/eca

Issues similares

Más issues de Clojure

Recibe los nuevos issues en tu correo

Un resumen breve de issues de GitHub para principiantes.