Add LLMTR to the free-provider directory

Abierto
  • ¿Algo no está bien?
#7,954 1 comentario 1 reacción 0 asignados Ver en GitHub

Los mantenedores suelen responder en 1 día

@knowhycodata ya está trabajando en esto.

Desde el 29/10/2026.

  • #7955 de @knowhycodata — abierto

Evaluación

Dificultad
3/5
Tiempo estimado
1-2 días
Aptitud para principiantes
28/100
Tipo de issue
Nueva funcionalidad
Claridad
Bien especificado
Estado de actividad
Activo
Stack tecnológico
typescript
Área
ai, backend

Línea de trabajo

Read the existing Runcrate directory entry and its focused test, referenced as #7726, then inspect the provider adapter directory and its tests. Add an LLMTR directory row with the stated base URL, docs and key links, and the six listed zero-price chat models; preserve live discovery for other models. Done means the row and test cover the requested catalog behavior. The author says a small PR is ready, so check for that work before starting.

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

Descripción

enhancement provider
Area

Provider adapters

What are you trying to accomplish?

Route Codex traffic through LLMTR, an OpenAI-compatible gateway at https://llmtr.com/v1. Disclosure: I maintain LLMTR (Knowhy, Istanbul). Chat uses Authorization: Bearer with a key starting with llmtr-; GET /v1/models is public.

What prevents this today?

opencodex has no LLMTR row. A custom OpenAI-compatible provider works, but each user has to look up the base URL, the key dashboard, and which models are actually free.

What should OpenCodex do?

List LLMTR in the free-provider directory, the same way Runcrate landed in #7726: base URL https://llmtr.com/v1, docs https://llmtr.com/docs/gateway/chat-completions/, keys at https://llmtr.com/dashboard/api-keys. Seed the standing zero-price chat models and keep live discovery for the rest of the catalog.

As of 2026-10-29, GET /v1/models returns 382 rows. These six chat ids have a zero list price with no end date:

  • qwen/qwen3.8-27b-free, nvidia/nemotron-3-ultra-550b-a55b, nvidia/nemotron-3.5-super-vl-free (daily quota)
  • openai/gpt-oss-safeguard-20b, agnes/agnes-2.5-flash, agnes/agnes-3.0-flash

A key is still required. Other zero-price rows are dated promotions or bring-your-own-key routes; everything else is paid, including llmtr/qwen3-8-27b. Pricing: https://llmtr.com/pricing. Where a model accepts reasoning, the field is a boolean.

This should be a directory row, not a canonical preset: most of the catalog is paid and LLMTR routes third-party models.

Example usage or interface
ocx provider add llmtr --adapter openai-chat --base-url https://llmtr.com/v1 \
  --auth-mode key --default-model qwen/qwen3.8-27b-free --set-default
printf '%s' "$LLMTR_API_KEY" | ocx account add-key llmtr
curl -sS http://127.0.0.1:10100/v1/chat/completions -H 'Content-Type: application/json' \
  -d '{"model":"qwen/qwen3.8-27b-free","messages":[{"role":"user","content":"Say hi in one word."}]}'
Alternatives or workarounds

A hand-written OpenAI-compatible provider with the same base URL.

Additional context

I couldn't find an existing LLMTR issue. I have a small PR ready against dev that adds the row and a focused test, and will link it here.

Checks
  • I searched existing issues and documentation.
  • This request describes a concrete OpenCodex workflow rather than merely naming a desired technology.
  • I removed secrets and personal data.
Lenguaje dominante
TypeScript
Estrellas
16.9k
Forks
1.3k
Merge medio
16 h 44 min
PR fusionados (30 d)
689

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 lidge-jun/opencodex

Todos los issues de lidge-jun/opencodex

Issues similares

Más issues de TypeScript

Recibe los nuevos issues en tu correo

Un resumen breve de issues de GitHub para principiantes.