[Bug]: App crash using OpenRouter for models like Qwen-2.5 72b

[enyst]

Is there an existing issue for the same bug?

• I have checked the existing issues.

Describe the bug and reproduction steps

If you choose an LLM like openrouter/qwen/qwen-2.5-72b-instruct, it fails with:

        ^^^^^^^^^^^^^^^^^^^^^^
  File "/Users/enyst/repos/odie/openhands/llm/llm.py", line 87, in __init__
    self.model_info = litellm.get_model_info(self.config.model)
                      ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "/Users/enyst/repos/odie/.venv/lib/python3.12/site-packages/litellm/utils.py", line 4983, in get_model_info
    raise Exception(

ERROR:root:<class 'Exception'>: This model isn't mapped yet. model=openrouter/qwen/qwen-2.5-72b-instruct, custom_llm_provider=openrouter. Add it here - https://github.com/BerriAI/litellm/blob/main/model_prices_and_context_window.json.

Normally, this shouldn't crash, just not have model_info going forward.

OpenHands Installation

Development workflow

OpenHands Version

No response

Operating System

None

Logs, Errors, Screenshots, and Additional Context

No response