openai-proxy does not support custom reasoning and embedding models

[DatCaptainHorse]

Describe the bug
When self-hosting and trying to use OpenAI server with custom models (Qwen3) loaded, reasoning cannot be enabled in Letta application and embedding model cannot be changed.

Please describe your setup

• How are you running Letta?
  • Docker Podman
• Describe your setup
  • Linux, CachyOS (Arch-based distro)
  • Running letta self-hosted in container

Screenshots

Additional context

• What model you are using?
  • Qwen3 14B (self-hosted)

Agent File (optional)
It's just memory-agent template.

---

If you're not using OpenAI, please provide additional information on your local LLM setup:

Local LLM details

If you are trying to run Letta with local LLMs, please provide the following information:

• The exact model you're trying to use (e.g. dolphin-2.1-mistral-7b.Q6_K.gguf)
  • Qwen3 14B
• The local LLM backend you are using (web UI? LM Studio?)
  • OpenArc (openai compatible)
• Your hardware for the local LLM backend (local computer? operating system? remote RunPod?)
  • local workstation with Intel Arc GPUs.